Research Article

Mitochondrial volume fraction and translation duration impact mitochondrial mRNA localization and protein synthesis

Department of Chemistry and Biochemistry, University of California San Diego, United States
Department of Developmental and Cell Biology and Center for Complex Biological Systems, University of California Irvine, United States
Division of Biological Science, Graduate School of Science, Nagoya University, Japan
Department of Anatomy and Structural Biology, Albert Einstein College of Medicine, United States
Gruss-Lipper Biophotonics Center, Albert Einstein College of Medicine, United States
Department of Neuroscience, Albert Einstein College of Medicine, United States
Janelia Research Campus, Howard Hughes Medical Institute, United States

Aug 7, 2020

https://doi.org/10.7554/eLife.57814

Version of Record

Accepted for publication after peer review and revision.

Download
Cite
Share
CommentOpen annotations (there are currently 0 annotations on this page).

Version of Record published: August 7, 2020 (This version)
Accepted: July 23, 2020
Received: April 13, 2020

1. Of interest
Insights into cargo sorting by SNX32 and its role in neurite outgrowth

Jini Sugatha, Amulya Priya ... Sunando Datta

Research Article Updated May 26, 2023
Further reading

Article
Figures and data
Abstract
Introduction
Results
Discussion
Materials and methods
Data availability
References
Decision letter
Author response
Article and author information
Metrics

Abstract

Mitochondria are dynamic organelles that must precisely control their protein composition according to cellular energy demand. Although nuclear-encoded mRNAs can be localized to the mitochondrial surface, the importance of this localization is unclear. As yeast switch to respiratory metabolism, there is an increase in the fraction of the cytoplasm that is mitochondrial. Our data point to this change in mitochondrial volume fraction increasing the localization of certain nuclear-encoded mRNAs to the surface of the mitochondria. We show that mitochondrial mRNA localization is necessary and sufficient to increase protein production to levels required during respiratory growth. Furthermore, we find that ribosome stalling impacts mRNA sensitivity to mitochondrial volume fraction and counterintuitively leads to enhanced protein synthesis by increasing mRNA localization to mitochondria. This points to a mechanism by which cells are able to use translation elongation and the geometric constraints of the cell to fine-tune organelle-specific gene expression through mRNA localization.

Introduction

Mitochondria are essential cellular organelles that are key sources of ATP generation via oxidative phosphorylation as well as the assembly of iron-sulfur clusters and many other catabolic and anabolic reactions (Attardi and Schatz, 1988). To support mitochondrial function, proteins from hundreds of nuclear genes are imported into mitochondria from the cytoplasm (Morgenstern et al., 2017). This has to be coordinated with the gene expression of the mitochondrial genome, which in Saccharomyces cerevisiae contains 13 genes (Borst and Grivell, 1978). While cells can generate ATP through mitochondrial oxidative phosphorylation, they can also use glycolysis as an alternative means of generating ATP. S. cerevisiae are Crabtree-positive yeast and will actively repress respiration and the use of alternative carbon sources in conditions in which the fermentable carbon source glucose is present (De Deken, 1966). This seems counterintuitive as the yield of ATP per glucose molecule is much higher in respiration compared to fermentation, but it is thought that fermentation allows higher fluxes of metabolite processing, leading to faster growth (van Dijken et al., 1993). Yet as cells run out of glucose they must switch their primary ATP generation source from fermentation to respiration. This metabolic change is known to dramatically change the mitochondrial morphology (Egner et al., 2002). The protein content of yeast mitochondria also shows dynamic changes in response to shifting cellular energy demands (Morgenstern et al., 2017; Paulo et al., 2016). The HAP complex is known to play an important role in the transcriptional upregulation of mitochondrial biogenesis upon a shift to non-fermentable carbon sources (Buschlen et al., 2003). Translational regulation has also been found to be important in the control of mitochondrial gene expression as oxidative phosphorylation protein coding mRNAs gradually increase their protein synthesis as the growth environment changes from fermentative growth to respiratory conditions (Couvillion et al., 2016). mRNA localization is a means to post-transcriptionally regulate gene expression at both a temporal and spatial level (Martin and Ephrussi, 2009). In the 1970s, electron microscopy analysis found that cytoplasmic ribosomes can be localized along the mitochondrial outer membrane (Kellems et al., 1974). Recent microarray and RNA-seq analyses of biochemically fractionated mitochondrial membranes and fluorescent microscopy analysis have identified subsets of nuclear-encoded mRNAs that are mitochondrially localized (Fazal et al., 2019; Gadir et al., 2011; Garcia et al., 2007; Marc et al., 2002; Saint-Georges et al., 2008; Williams et al., 2014). It has been shown that both the 3′ UTR and coding regions, primarily through mitochondrial targeting sequences (MTSs), contribute to mitochondrial localization. One class of localized mRNAs (Class I) was shown to be dependent on the Puf3 RNA-binging protein through binding motifs in the 3′ UTR (Saint-Georges et al., 2008). Another class of mRNAs was localized to the mitochondria independently of Puf3 (Class II). Many of the localized mRNAs show reduced association upon polysome dissociation through EDTA or puromycin treatment, implicating translation as a necessary factor for mRNA localization (Eliyahu et al., 2010; Fazal et al., 2019). The mitochondrial translocase of the outer membrane (TOM) complex has been shown to impact mRNA localization through interaction with the nascent MTS (Eliyahu et al., 2010), while the outer-membrane protein OM14 has been shown to be a mitochondrial receptor for the ribosome nascent-chain-associated complex (NAC) (Lesnik et al., 2014). Isolating mitochondrially localized ribosomes to perform proximity-specific ribosome profiling revealed that many mitochondrial inner-membrane protein mRNAs are co-translationally targeted to the mitochondria (Williams et al., 2014). These observations have suggested a mechanism of co-translational protein import into mitochondria for a subset of nuclear-encoded mitochondrial mRNAs.

While mRNA localization is a way to control gene expression and there is strong evidence for the localization of mRNAs to the mitochondria, the use of this mRNA localization to alter the composition of mitochondria in different environmental conditions has not been explored. Furthermore, though mitochondrial biogenesis is thought to be transcriptionally regulated in relation to the metabolic needs of the cell, how these changing mitochondrial dynamics may also directly impact mRNA localization and protein synthesis has not been investigated. Here we report that interactions between mitochondria and mRNA/nascent-peptide (MTS) complexes can be altered by both the kinetics of protein synthesis and the fraction of cytoplasm that is mitochondrial, leading to condition-dependent mitochondrial mRNA localization during respiratory conditions. This localization subsequently leads to enhanced protein expression for these condition-specific localized mRNAs during respiratory conditions.

Results

mRNA association with mitochondria differs between fermentative and respiratory conditions

To explore how mRNA localization is impacted by the metabolic state of the cell and changing mitochondrial morphologies, we developed a methodology to quantify mRNA location and mitochondrial 3D structure in living cells. To do this, we visualized mitochondria using the matrix marker Su9-mCherry and single-molecule mRNAs with the MS2-MCP system every 3 seconds in a microfluidic device (Figure 1A; Figure 1—figure supplement 1). We reconstructed and analyzed the spatial relationship between the mRNAs and mitochondria using the ImageJ plugin Trackmate (Tinevez et al., 2017) and MitoGraph V2.0, which we previously developed to reconstruct 3D mitochondria based on matrix marker fluorescent protein intensity (Rafelski et al., 2012; Viana et al., 2015; Figure 1A). We measured the distance between mRNA and mitochondria by finding the closest meshed surface area of the mitochondrial matrix (Figure 1—figure supplement 2; Video 1; Materials and methods).

Figure 1 with 2 supplements see all

Download asset Open asset

mRNA association with mitochondria differs between fermentative and respiratory conditions.

(A) Experimental setup for live imaging. Mitochondria were visualized by Su9-mCherry and mRNAs were visualized by the single molecule MS2-MCP tethering system. Z-stacks were taken within 1.5 s for each individual channel, and multiple Z-stacks were merged into a series. (Right top) Z-projected image of a live cell. Cyan: mRNA, magenta: matrix. Scale bar, 2 µm. (Right bottom) Reconstructed mRNA and mitochondria. Yellow: mRNA, blue: mitochondria. (B) The proportion of mitochondrial associated mRNA per cell (n > 27) of the different mRNA species in fermentative and respiratory conditions. Error bar represents standard error of the mean (s.e.m.). (C) *ATP3* mRNA and *TIM50* mRNA expression number per cell. MCP-GFP foci were counted per cell (n > 27). (D) Atp3p-GFP and Tim50p-GFP fusion protein expression level using GFP fluorescent intensity per cell. (C)-(D) Statistical significance was assessed by Mann–Whitney U-test (****p<0.0001; ***p<0.001; **p<0.01; ns, no significant difference). (E) Atp3p-GFP and Tim50p-GFP fusion protein expression level using western blotting with anti-GFP antibody. Each dot represents an independent experiment. Error indicates standard deviation of three independent experiments.

Video 1

Download asset

posterframe for video — Z-projection of live imaging of mRNA and mitochondria in cycloheximide and 1,10-phenanthroline treated cells.

*TIM50* and *TOM22* mRNA were visualized 10 minuetes after the addition of cycloheximide and 1,10-phenanthroline in fermentative condition. (Left) *TIM50* mRNA (Right) *TOM22* mRNA. Cyan: mRNA, magenta: matrix.

We first analyzed three different mRNAs (Saint-Georges et al., 2008; Williams et al., 2014). Two mRNAs have previously been found to be mitochondrially localized and contain a mitochondrial targeting sequence (MTS): ATP3 mRNA, which encodes the gamma subunit of ATP synthase, and TIM50, which encodes a component of the inner membrane translocase. The third mRNA, TOM22, encodes an outer-membrane translocase that does not contain an MTS and has previously been found to be predominantly diffusely localized (Gadir et al., 2011; Garcia et al., 2010; Williams et al., 2014). During fermentative growth, we observed TIM50 mRNA to be strongly associated with the mitochondria, whereas TOM22 showed low association consistent with previous studies (Figure 1B; Video 2; Materials and methods). Even though ATP3 had previously been categorized as a mitochondrially localized mRNA (Gadir et al., 2011; Saint-Georges et al., 2008), we unexpectedly found this to be condition-dependent as it has low association with mitochondria, similar to TOM22, in fermentative conditions. However, during respiratory conditions it strongly shifted towards association with the mitochondrial surface, in a manner more similar to TIM50 (Figure 1B). This means that nuclear-encoded mitochondrial mRNAs do not have to be solely mitochondrially localized or diffusely localized; instead, they can show a switch-like behavior depending on the metabolic state of the cell.

Video 2

Download asset

There are large changes in mitochondrial composition as yeast switch from fermentative to respiratory metabolism (Morgenstern et al., 2017). To investigate whether there may be a relationship between mRNA localization and gene expression, we measured mRNA levels via number of mRNAs per single cell and protein levels via both single-cell measurements and bulk assays in both fermentative and respiratory conditions. For the condition-dependent ATP3 mRNA, we found that protein levels increased 4-fold, whereas mRNA levels increased less than 2-fold when cells were grown in respiratory versus fermentative conditions. TIM50 mRNA, which is constitutively localized to the mitochondria under both conditions, showed no change in protein or mRNA levels in respiratory conditions (Figure 1C–E). This suggests a relationship between mRNA localization to the mitochondria and protein production.

Relationship of mRNA localization to mitochondrial volume fraction

As yeast cells shift to respiratory conditions, mitochondrial biogenesis increases the mitochondrial volume while the cell cytoplasmic volume decreases, thus leading to an increase in the mitochondrial volume fraction in respiratory conditions (Egner et al., 2002; Figure 2A; Figure 2—figure supplement 1). While ATP3 mRNA showed a strong condition-dependent localization, TIM50 and TOM22 mRNAs also showed modestly increased mitochondrial association during respiratory conditions (Figure 1B). We wondered what impact the reduction in the availability of free cytoplasmic space due to mitochondrial expansion had on mRNA co-localization, especially for TOM22, which is not known to bind to the mitochondria. To test this, we quantified both the mitochondrial localization of each mRNA and changes in mitochondrial volume fraction at a single-cell level. We found that TOM22 showed a linear increase in co-localization that was directly proportional to mitochondrial volume fraction (Figure 2B). We also found that ATP3 mRNA was more sensitive to mitochondrial volume fraction than TIM50 and TOM22. This sensitivity was independent of nutrients, as fermentative yeast cells showed a larger increase in ATP3 localization as the mitochondrial volume fraction increased (Figure 2B). At the lowest mitochondrial volume fractions, ATP3 localization was similar to the diffusely localized TOM22 mRNA, whereas at the highest volume fractions, its localization was close to the mitochondrially localized mRNA TIM50. This suggests that increased mitochondrial volume fraction in turn increases ATP3 mRNA localization to mitochondria.

Figure 2 with 5 supplements see all

Download asset Open asset

Mitochondrial volume fraction correlates with mRNA localization.

(A) Snapshot of reconstructed mitochondrial surface (blue) and *TIM50* mRNA foci (yellow) in fermentative and respiratory conditions. Scale bar, 1 µm. (B) Relationship between the mitochondrial volume fraction and the proportion of mRNA localization to mitochondria. Trend line was depicted according to the best linear fit of the proportion of localization and mitochondrial volume fraction of single cells in each condition of different mRNAs (n > 27). Dotted red line marks the difference between fermentative and respiratory conditions for mitochondria volume fraction. Gray region surrounding the trend lines represents the 95% confidence interval (CI) for each line. (C) Schematic of *in silico* experiment. (Top) Brownian particle distribution indicates that mRNA localization rate correlates with mitochondrial volume fraction. (Bottom) Hypothetical thermodynamic equilibrium of binding of mRNA to mitochondria. [mRNA_localized], mitochondrial localized mRNA; [mRNA_free], free diffusing mRNA; [mito_surface], mitochondrial surface where mRNA can bind. (D) Mitochondrial volume fraction and mRNA localization have stoichiometric correlation. Relationship of proportion of mRNA localization to mitochondria and mitochondrial volume fraction from mathematical modeling was plotted. Yellow line represents linearly fitted line for *TOM22* mRNA. Green and blue lines were plotted through equilibrium constant of 2.4K₀ and 8.8K₀, respectively, as described in Materials and methods. (E) In glucose conditions, *sch9*∆ and *reg1*∆ mutant strains as well as chloramphenicol addition exhibit higher mitochondrial volume fraction and increase the localization of *ATP3* mRNA to the mitochondria (n > 27). Predicted relationship of proportion of mRNA localization to mitochondria and mitochondrial volume fraction from mathematical modeling of WT strains was plotted as a green dotted line. Mean values of each axis for WT, mutant cells, and cells with chloramphenicol addition were plotted as crosses. Error bar represents s.e.m.

To further test our hypothesis that mRNA localization is regulated by mitochondrial volume fraction, we designed in silico experiments based on our experimentally measured cell and mitochondrial boundaries and used a mathematical model to investigate how particles of varying affinities would co-localize with mitochondria (Figure 2C). We were able to recapitulate the behavior of TOM22 via a model of an ideal Brownian particle with no affinity for the mitochondria, which showed linearly correlated localization with mitochondrial volume fraction (Figure 2C, equation (1); Figure 2D; Figure 2—figure supplement 2A). We then set up a simple equilibrium equation where the baseline equilibrium constant, K₀, was set by a freely diffusing particle like TOM22 and multiplied by the affinity, A, of the particle for the mitochondria, thus giving K = AK₀ (Figure 2C, equation (2); Materials and methods). As the value of A increased in the simulation, the proportion of mitochondrial localization of the mRNA for a given mitochondrial volume fraction increased as well (Figure 2—figure supplement 2B). We then applied this relationship to estimate the experimental values of A to be 2.4 and 8.8 for ATP3 and TIM50, respectively (Figure 2D; Figure 2—figure supplement 2C). Notably, mitochondrial volume fraction correlates with mRNA localization, but cell volume and mitochondrial volume alone did not show any correlation through fermentative and respiratory conditions (Figure 2—figure supplement 2C–E). Interestingly, this simple mathematical relationship also recapitulates the shape of the curves in Figure 2B, suggesting that the apparent shift in association that occurs during the switch from fermentative to respiratory conditions may be primarily a result of the combination of the mitochondrial volume fraction and the strength of mRNA-specific association and, therefore, not due to the difference in growth condition or to other mechanisms.

To further test this model, we sought to manipulate mitochondrial volume fraction in a multitude of ways. We first tested the impact of increased mitochondrial biogenesis on mRNA localization while maintaining the nutrient conditions of the cell. Reg1 is a protein necessary for glucose repression in yeast, and reg1∆ cells exhibit increased mitochondrial function in glucose media (Adachi et al., 2017; Hübscher et al., 2016). Supporting our model, we observed increased mitochondrial volume fraction in rich glucose conditions in reg1∆ mutant cells as well as increased ATP3 mRNA localization to the mitochondria (Figure 2E; Figure 2—figure supplements 3A–D and 4). Additionally, the antibiotic chloramphenicol is known to decrease mitochondrial protein synthesis while having no effect on cytoplasmic translation and has been found to increase the fraction of mitochondrial protein in Neurospora crassa and mitochondrial volume in Tetrahymena pyriformis (Brody, 1992; Gleason et al., 1975). We sought to test whether chloramphenicol would alter the mitochondria volume fraction in yeast during rich glucose conditions when mitochondrial respiration is not required. We observed an ~70% increase in mitochondria volume fraction in yeast cells treated with chloramphenicol and, simultaneously, a ~ 50% increase in ATP3 mRNA mitochondrial association during vegetative conditions (Figure 2E; Figure 2—figure supplements 3A–D and 4). These results support the hypothesis that mitochondrial volume fraction is the important variable for ATP3 mRNA localization and not a factor related to respiration, as chloramphenicol decreases respiratory function (Williamson et al., 1971) while increasing mitochondrial volume fraction and ATP3 localization.

Finally, as mitochondrial volume fraction can be increased by either increasing mitochondrial volume or decreasing cytoplasmic volume, we tested whether decreasing cytoplasmic volume could increase ATP3 mRNA mitochondrial association. sch9∆ was previously found to be one of the smallest strains from a genome-wide screen for cell size (Jorgensen et al., 2002). We found that ATP3 mRNAs showed increased mitochondrial association in sch9∆, though from initial analysis this was not accompanied by an increase in mitochondrial volume fraction. While imaging these cells, we noticed a large increase in the relative vacuole in sch9∆ cells; the vacuole volume fraction was ~6 x higher in sch9∆ cells versus WT cells. As the vacuole restricts the accessible cytoplasmic volume, we recalculated the mitochondrial volume fraction relative to the accessible cytosol and found that sch9∆ cells have a significant increase in mitochondrial volume fraction after this correction (Figure 2—figure supplements 3E–F and 4). With this corrected mitochondrial volume fraction, we found that, similar to reg1∆ and chloramphenicol-treated cells, sch9∆ cells had increased localization of ATP3 mRNA to the mitochondria that was not significantly different either from our experimental measurements relating mitochondrial volume fraction and mRNA association in WT cells or from what would be predicted from our mRNA localization simulation based solely on changing mitochondrial volume fraction (Figure 2E; Figure 2—figure supplements 3G–I and 4). These results show that the relationship between mitochondrial volume fraction and mRNA localization holds across highly varied perturbations, including nutritional (glucose/glycerol), genetic (reg1∆, sch9∆), and pharmacological (chloramphenicol).

An alternative possibility is that expression of the import machinery or the NAC receptor correlate with mitochondrial volume fraction and that these are what is important for the changes in ATP3 mRNA localization. If this is the case, we expect that the expression of these genes should increase across all perturbations that have higher mitochondrial volume fraction and increased ATP3 localization. To test this, we investigated the expression of potential known receptors in fermentative and respiratory conditions using published proteomics data (Morgenstern et al., 2017). Surprisingly, most of these potential receptors, including the import machinery, TOM/SAM complex, do not show a difference in expression between respiratory and fermentative conditions (Figure 5—source data 1). The outer membrane receptors OM14 and OM45 did have a significant 7 to 9-fold increase in expression when mitochondrial volume fraction is expanded. To further explore the role of these outer membrane proteins in mRNA localization we measured their expression in reg1∆ and sch9∆ mutant strains, which have increased mitochondrial volume fraction and ATP3 localization, but are grown in glucose media (Figure 2E). While we are able to recapitulate the large increase in expression of OM14 and OM45 in respiratory conditions, we do not find a significant increase in the expression of these RNAs in reg1∆ and sch9∆ mutant strains (Figure 2—figure supplement 5). Overall, these data point to the importance of increasing mitochondrial volume fraction effectively shrinking the cytoplasm and favoring random interactions with the mitochondrial membrane, which would result in altered association based on binding affinities of the components involved. Furthermore, mRNA association with mitochondria can be tuneable to permit a switch-like transition in mitochondrial localization, as is seen for ATP3 mRNA (Figure 1B), due to a nutrient-induced change in mitochondrial volume fraction.

Translation regulates mRNA localization

Given these results, we wanted to delve further into the mechanism of this varying localization and protein production. Even though the mitochondrial protein import machinery is well described, an ER-like signal recognition particle dependent mechanism of co-translational protein import has not been identified for the mitochondria (Golani-Armon and Arava, 2016; Reid and Nicchitta, 2015). However, a series of biochemistry and microscopy analysis showed that some nuclear-encoded mitochondrial protein mRNAs are translated on the mitochondrial surface (Gadir et al., 2011; Garcia et al., 2007; Gold et al., 2017; Kellems et al., 1974; Marc et al., 2002; Saint-Georges et al., 2008; Williams et al., 2014). We therefore investigated the effects of the MTS and of protein translation on mRNA association with mitochondria. We remove or replaced the MTS of Tim50p with an ER-localization signal or introduced an ER-targeting signal at the N-terminus of Tom22p (Wu et al., 2016). Even though TIM50 mRNA was associated with mitochondria, ER-TIM50, TOM22, ER-TOM22, and TIM50-∆MTS mRNAs were not associated with mitochondria (Figure 3A, Figure 3—figure supplement 1), indicating that the TIM50 MTS is necessary to recruit this mRNA to mitochondria. To further support the role of the MTS in mRNA localization, we tested whether reducing ribosome-nascent chain association by using the translation initiation inhibitor lactimidomycin (LTM) would affect mRNA localization. We found that TIM50 mRNA in all conditions and ATP3 mRNA in respiratory conditions decreased localization to the mitochondrial surface upon LTM addition, while ATP3 mRNA in fermentative conditions showed only minimal changes in localization upon LTM addition (Figure 3A). These results suggest that the actively translating ribosome drives mRNA localization to mitochondria through production of the nascent N-terminal MTS.

Figure 3 with 2 supplements see all

Download asset Open asset

Increased protein synthesis and mRNA localization is regulated by the downstream coding sequence.

(A) An ER-localization signal and translation inhibitor drugs alter the proportion of mitochondrial associated mRNA per cell of the strains in Figure 1B (n > 20). LTM indicates 50 mM lactimidomycin for 20 min. Error bar represents s.e.m. Statistical significance compared with control (value in Figure 1B, red dotted line) was assessed by Mann–Whitney U-test (****p<0.0001; *p<0.05; ns, no significant difference). (B) Schematic of chimeric reporter genes for swapping of MTS (1-100aa) and CDS (101aa-) between *TIM50* and *ATP3*. Chimeric genes were conjugated with either *flagGFP* or *flagiRFP*. (C) Protein expression from reporter genes with flagGFP depicted in (B). Growth ‘F’ represents fermentative and ‘R’ represents respiratory conditions. Tub1p was used as internal loading control. Protein expression ratio between respiratory and fermentative conditions is shown in the bottom row. Error indicates standard deviation of three independent experiments. (D) The proportion of mitochondrial associated mRNA per cell of reporter mRNAs *with flagiRFP* in fermentative conditions (n > 34). Error bar represents s.e.m. Statistical significance was assessed by Mann–Whitney U-test (*p<0.05; ns, no significant difference).

From our simulation, we were able to recapitulate our experimental observation that TIM50 mRNA has higher affinity for the mitochondria than ATP3 mRNA, making its localization less dependent on mitochondrial volume fraction. As the MTS was necessary for localization to the mitochondria, our initial hypothesis was that Tim50p MTS has a higher affinity for the mitochondria than the Atp3p MTS, causing the differences in mRNA affinities. To test this hypothesis, we designed chimeric reporter genes wherein we swapped the MTS sequences between Tim50p and Atp3p under TIM50 promoter control (Figure 3B). Surprisingly, we found that the downstream coding sequence (CDS) was what differentiated TIM50 from ATP3, not the MTS. When the reporter gene contained the TIM50-CDS, it showed uniform protein production in fermentative versus respiratory conditions, independent of which MTS was present (Figure 3C, lane 1 vs. 2 and lane 5 vs. 6). However, when the reporter gene contained the ATP3-CDS, it showed decreased protein production in fermentative conditions (Figure 3C, lane 3 vs. 4 and lane 7 vs. 8). Similarly, the reporter genes that harbored the ATP3-CDS also showed decreased proportion of mitochondrial associated mRNA in fermentative but not in respiratory conditions (Figure 3D, Figure 3—figure supplement 2). These experiments suggest that the TIM50 and ATP3 MTS have similar affinities for the mitochondria and, importantly, what drives the condition-specific differences in mRNA localization and protein production between these mRNAs is encoded in the downstream CDS.

Ribosome stalling is important for constitutive mitochondrial localization

Our model proposes that the reason ATP3 mRNA increases localization in respiratory conditions is that the increased mitochondrial volume fraction increases the probability that the nascent MTS will interact with the mitochondrial surface. If the ATP3 and TIM50 MTS have similar affinity for the mitochondria, we hypothesized that the reason TIM50 has higher mitochondrial association at lower mitochondrial volume fractions is because the downstream CDS increases the chance of association between mitochondria and MTS, possibly by slowed translation elongation. Upon further examination, we found that the TIM50 CDS has a run of 7 consecutive proline codons approximately 60 amino acids downstream of the MTS. Polyproline stretches have been shown to mediate ribosome stalling, and, when we investigated a ribosome profiling data set, we found that ribosomes accumulate at this polyproline stretch during fermentative conditions (Zid and O’Shea, 2014; Figure 4—figure supplement 1). This suggests a possible mechanism, similar to what has been seen for SRP recognition, by which local slowdown of ribosomes increases the chance that the mitochondria will recognize the TIM50 MTS and consequently promote its association with the mitochondrial surface (Pechmann et al., 2014; Zhang and Shan, 2012). To test this, we deleted these polyproline residues and found this caused TIM50 to be more sensitive to environmental conditions as it reduced both protein synthesis and mRNA localization of TIM50 during fermentative conditions (Figure 4A–D). In contrast, the ATP3 coding sequence does not have any obvious strong ribosome stalling sequence. This suggests that ATP3 mRNA localization and protein synthesis are regulated solely in a manner dependent on mitochondrial volume fraction. If this is true, artificially slowing the ribosomes in fermentative conditions should drive ATP3 mRNA to become mitochondrially localized. To test this hypothesis, we inserted a run of 7 consecutive proline codons at 100 amino acids downstream of the start codon and analyzed protein expression in fermentative conditions (Figure 4A, right). We observed that protein production is increased 1.6-fold (Figure 4E) and mRNA localization is increased as well (Figure 4F). An alternative explanation is that the RNA sequence encoding the polyprolines is important for localization to the mitochondria and not ribosome stalling due to the encoded polyprolines. We believe this is not the case as translation initiation is key for the localization of TIM50 mRNA to the mitochondria (Figure 3A), and that upon lactimidomycin treatment TIM50 and ATP3 have indistinguishable localization phenotypes even though TIM50 mRNAs contain the RNA sequence for seven consecutive polyprolines and ATP3 does not, arguing against the RNA sequence directly having a role in the differential localization between TIM50 and ATP3. Overall these results further suggest that ribosome stalling leads to mRNA localization to mitochondria and increased protein production.

Figure 4 with 1 supplement see all

Download asset Open asset

Decreased translational elongation localizes mRNA to mitochondria.

(A) Schematic of deletion of polyproline sequence from *TIM50-flagGFP/flagiRFP* reporter gene and insertion of polyproline sequence into *ATP3- flagGFP/flagiRFP* reporter gene. These constructs are called *TIM50*-P7∆ and *ATP3*+P7. (B) Protein expression from reporter genes *TIM50-flagGFP* and *TIM50*-P7∆*-flagGFP*. Growth ‘F’ and ‘R’ correspond to fermentative and respiratory conditions, respectively. Tub1p was used as internal loading control. Protein expression ratio between respiratory and fermentative conditions is shown in the bottom row. Error indicates standard deviation of three independent experiments. (**C, D**) The proportion of the mitochondrial associated mRNA per cell (n > 20) of the reporter mRNAs *TIM50-flagiRFP* and *TIM50*-P7∆*-flagiRFP* in fermentative (C) and respiratory (D) conditions. (E) Protein expression from reporter genes *ATP3-flagGFP* and *ATP3*+P7-flagGFP in respiratory condition. Tub1p was used as internal loading control. Protein expression ratio between the reporter genes is shown in the bottom row. Error indicates standard deviation of three independent experiments. (F) The proportion of mitochondrial associated mRNA per cell (n > 29) of the reporter mRNAs *ATP3-flagiRFP* and *ATP3*+P7∆*-flagiRFP* in fermentative conditions.

To further explore whether slowing translation elongation stabilizes the mRNA-ribosome complex with the MTS, thereby giving it more time to associate with mitochondria, we measured mitochondrial mRNA localization following the addition of the translation elongation inhibitor cycloheximide (CHX) (Figure 5A). As our hypothesis predicted, we observed a 3-fold increase in the association of ATP3 mRNA with mitochondria during fermentative conditions but no increase in the case of TOM22 mRNA (Figure 5A). Interestingly, CHX treatment only slightly increased TIM50 mRNA localization. This potentially suggests that a portion of these mRNAs are incompetent for binding, perhaps due to nuclear localization or being in the midst of degradation, and that the majority of competent TIM50 mRNAs were already localized to the mitochondria.

Figure 5 with 2 supplements see all

Download asset Open asset

Decreased translational elongation localizes mRNA to mitochondria.

(A) Translational inhibitor drugs alter the proportion of mitochondrial-associated mRNA per cell (n > 43) of the strains in Figure 1C. CHX+PH indicates 100 µg/mL cycloheximide and 200 µg/mL 1,10-Phenanthroline for 10 min in fermentative condition. Error bar represents s.e.m. (B) Percent distribution of *TIM50* and *ATP3* mRNA mitochondrial localization behavior in annotated mitochondria protein encoding genes. We re-summarized and classified TIM50-type mRNA as enriched at the mitochondria regardless of CHX addition and ATP3-type mRNA as enriched at the mitochondria only upon CHX addition using proximity ribosome profiling data of annotated mitochondria protein encoding genes (Mitop2) (Elstner et al., 2009; Williams et al., 2014). All other mRNAs were considered not localized. (C)-(D) The proportion of the mitochondrial associated mRNA per cell (n > 16) of MS2-tagged endogenous *ATP2* and *ATP7* mRNAs in fermentative, respiratory, and CHX-treated for 10 min in fermentative conditions. Error bar represents s.e.m. Statistical significance in this figure was assessed by Mann–Whitney U-test (****p<0.0001; ***p<0.001; **p<0.01; *p<0.05; ns, no significant difference). (E) Protein production is increased upon a switch from fermentative to respiratory conditions in ATP3-type genes. Cumulative distribution of protein production (ribosome reads) (Couvillion et al., 2016) of TIM50-type (n = 60) and ATP3-type genes (n = 44) were depicted with blue and green lines, respectively. Statistical significance in this figure was assessed by Student’s t-test. Source data of ribosome profiling used for the analysis are summarized in the Source data 1.

Figure 5—source data 1 Summary of transcriptomics and proteomics from previously published data. Mitochondrial enrichment of mRNA with and without addition of cycloheximide (Williams et al., 2014), MLR Classification (Saint-Georges et al., 2008), translational efficiency and transcript abundance from previous ribosome profiling dataset (Couvillion et al., 2016), and protein abundance from previous mass spectrometry dataset (Morgenstern et al., 2017) were summarized in the Table.: https://cdn.elifesciences.org/articles/57814/elife-57814-fig5-data1-v1.xlsx
Download elife-57814-fig5-data1-v1.xlsx

A previous study found that 130 of 551 annotated nuclear-encoded mitochondrial mRNAs are sensitive to translation elongation rate and become localized to the mitochondrial surface upon cycloheximide treatment, similar to ATP3 (Figure 5B; Williams et al., 2014). Interestingly, all of the ATP synthase subunits that are conserved from bacteria to eukaryotes are sensitive to cycloheximide, except for the ε subunit, ATP16, whereas all of the nonconserved subunits are insensitive to cycloheximide (Table 1). We wondered whether this sensitivity may be indicative of mRNAs that also switch their localization as the mitochondrial volume fraction increases during respiratory conditions (Table 1). We tested ATP2, the conserved β subunit of the ATP synthase, and found that it was similar to ATP3 in that it showed a large increase in localization according to volume fraction change upon a shift to respiratory conditions and was sensitive to cycloheximide (Figure 5C; Figure 5—figure supplement 1). However, the nonconserved subunit, ATP7, behaved more like TOM22 in that it was insensitive to cycloheximide and had a much smaller increase in mitochondrial localization in respiratory conditions than ATP2 or ATP3 (Figure 5D). We then wanted to more globally explore the connection between sensitivity to translation elongation and changes in gene expression during the metabolic shift from fermentation to respiration when mitochondrial volume fraction dramatically increases. We focused on Class II mRNAs that were found to be localized to the mitochondria during respiratory conditions independently of Puf3 (Saint-Georges et al., 2008) and subdivided these into ATP3-type mRNAs that were cycloheximide-sensitive in localization to mitochondria during fermentative conditions and TIM50-type that were constitutively localized to mitochondria in fermentative conditions (Williams et al., 2014). We used previously generated ribosome profiling data on yeast cells under glucose and glycerol conditions (Couvillion et al., 2016). From this data, we found that ATP3-type mRNAs had a significant increase in their protein productive capacity versus TIM50-type as they had >2 fold more ribosomes engaged in translation in the shift from glucose to glycerol (Figure 5E). The increase in protein production was caused by both a significant increase in translation efficiency and mRNA levels for ATP3-type mRNAs during respiratory conditions (Figure 5—figure supplement 2A,B). From an independent proteomics data set under glucose and glycerol conditions, we also found that ATP3-type genes had an increase in protein abundance during respiratory conditions versus TIM50-type genes (Figure 5—figure supplement 2C; Morgenstern et al., 2017). These results point to translational elongation sensitivity and condition-dependent localization being general strategies to fine-tune gene expression of certain nuclear-encoded mitochondrial genes.

Table 1

mRNAs of conserved ATP synthetase components are localized to mitochondria in respiratory and CHX conditions.

ATP complex gene conservation was obtained from published annotation (Rühle and Leister, 2015) and we obtained data of CHX-dependent mitochondrial localization from proximity ribosomal profiling data (Williams et al., 2014). The CHX-dependent localization data were consistent with our study. Respiratory-specific localization was determined in this study.

	Complex	Subunit	Gene	Mitochondrial localization		Respiratory specific localization
	Complex	Subunit	Gene	-CHX	+CHX	Respiratory specific localization
Conserved from bacteria to eukaryote	F1	α	ATP1	no	yes	yes
		β	ATP2	no	yes	yes
		γ	ATP3	no	yes	yes
		OSCP/δ	ATP5	no	yes	NA
		ε	ATP16	no	no	NA
	F0	b	ATP4	no	yes	NA
non-conserved	all the others			no	no	NA (ATP7: no)
Rühle and Leister, 2015				Williams et al., 2014		This study

mRNA localization to the mitochondria drives enhanced protein synthesis

Recently, it has been shown that there is active cytoplasmic translation on the mitochondrial surface in Drosophila and that a subset of nuclear-encoded mitochondrial proteins are translationally regulated by the localization of specific RNA-binding proteins to the mitochondrial surface (Zhang et al., 2016). We therefore hypothesized that mRNA localization to mitochondria may be a way to drive the coordinated increase in mitochondrial protein production observed in respiratory conditions. An alternative explanation is that increased translation drives more nascent protein production, which increases mRNA localization to the mitochondria. To directly test these two possibilities, we analyzed the effect of driving mRNA localization to mitochondria on protein expression. To accomplish this, we tethered reporter mRNAs to mitochondria by MS2 sequences. We inserted the MCP protein into the C-terminus of Tom20p and Tom70p, two well-characterized proteins on the outer mitochondrial membrane, and analyzed subsequent protein production (Figure 6A). We found that tethering TIM50-flag-GFP-12xMS2 and ATP3-flag-GFP-12xMS2 mRNA to the mitochondria was sufficient to upregulate protein production. Surprisingly, protein production was increased independent of the mRNA harboring an MTS, as an mRNA that contained flag-GFP with no mitochondrial coding sequences also showed increased protein production (Figure 6B–C). We then analyzed whether tethering to the ER might affect protein production by inserting the MCP protein into the C-terminus of Sec63p. We also saw increased protein production when mRNA was tethered to the ER, suggesting that the surface of both of these organelles may harbor the capacity for enhanced protein synthesis. In addition to increased protein expression, we also observed increased mRNA levels when mRNAs were tethered to the mitochondria. However, the ratio of protein to mRNA was much higher (Figure 6D, Figure 6—figure supplement 1A). We further observed a 3-fold increase in ribosome association rate from sucrose density fractionated mRNAs for mRNAs that were tethered to the mitochondria (Figure 6E, Figure 6—figure supplement 2), suggesting that translational efficiency is increased on the mitochondrial surface. To test whether localization to the mitochondria is necessary for optimal protein production during respiratory conditions, we reduced the localization of endogenous ATP3 and TIM50 mRNA to mitochondria by directing those mRNAs to the plasma membrane via insertion of a CaaX-tag to the C termini of MCP-GFP proteins during respiratory conditions (Yan et al., 2016). This caused a decrease in protein reporter levels of mitochondrial localized mRNAs, such as TIM50, ATP3, and ATP2, but not in non-mitochondrially localized mRNAs, such as TOM22 and GFP alone (Figure 6F, Figure 6—figure supplement 1B). We next investigated whether enhancing protein synthesis was essential for optimal cell growth. Cells in which ATP3 mRNA was anchored to the plasma membrane and away from the mitochondria in respiratory conditions showed a growth defect, whereas ER tethering of mRNAs, which does not impair protein synthesis, did not affect cell growth (Figure 6G; Figure 6—figure supplement 3). This suggests that localization of mRNA to mitochondria is important for optimal cell growth because it drives enhanced protein synthesis during respiratory conditions.

Figure 6 with 3 supplements see all

Download asset Open asset

mRNA localization to mitochondria enhances its translation.

(A) Schematic of artificial mRNA tethering to mitochondria using MS2-MCP system. mRNAs harboring a MS2 tandem sequence were tethered to mitochondria through C-terminus MCP tagged Tom70p or Tom20p. (B) Protein expression analysis of reporter mRNAs, which are tethered to mitochondria (lanes 2, 3) or ER (lane 4). Protein expression was analyzed using an anti-GFP antibody. Tub1p in the strain harboring flagGFP-12xMS2 reporter genes was used as an internal control. (C) Quantification of (B). Protein expression was normalized to the No-MCP strains. (D) Expression level of reporter mRNAs tethered to mitochondria or ER. mRNA expression was analyzed with RT-qPCR using primers for flagGFP. (E) Relative ribosome association rate of *flagGFP-12xMS2* expressed with MCP-GFP or TOM70-MCP was calculated by comparison between RNA levels from ribosome-bound and free fractions from sucrose gradient polysome fractionation. RNA levels were quantified via qPCR. (F) Protein expression analysis in anchored-away conditions. Protein expression levels in each strain in respiratory conditions were determined by western blotting using anti-flag antibody. Protein expression ratio between strains with and without CaaX is shown as a bar graph. (C)-(F), Each dot represents an independent experiment. Error bar represents standard deviation of three independent experiments. (G) Growth assay for plasma membrane localized mRNA, which consists of an integrated MS2 sequence into the 3’-UTR of genomic DNA. *TIM50* and *ATP3* mRNAs were anchored away to the plasma membrane using CaaX-tag harbored MCP-GFP proteins. Cell growth was tested on YPAD (fermentative) and YPAGE (respiratory) conditions at 30°C for 2 days and 3 days, respectively.

Discussion

During fluctuating environmental conditions, cells must be able to control gene expression in order to optimize fitness. We demonstrate that yeast cells can use the geometric constraints that arise from increased mitochondrial volume fraction during respiratory conditions to increase condition-dependent mitochondrial localization for a subset of nuclear-encoded mitochondrial mRNAs. We favor the hypothesis that the geometric constraints of mitochondrial volume fraction are an important factor impacting mRNA localization for two reasons: first, our simple mathematical model incorporating mitochondrial volume fraction and various binding affinities is able to recapitulate the mRNA localization effects we see in cells. Second, the relationship between mitochondrial volume fraction and mRNA localization holds across a multitude of experimental perturbations, which presumably impact mitochondrial function in very different ways. reg1∆ changes mitochondrial volume fraction exclusively by increasing mitochondrial volume, whereas chloramphenicol and a nutrient shift from glucose to glycerol media increase mitochondrial volume fraction by both increasing mitochondrial volume and decreasing cytoplasmic volume. While reg1∆ and glycerol media increase oxidative phosphorylation, chloramphenicol inhibits mitochondrial translation and respiratory function. Finally, sch9∆ has a reduced mitochondrial volume but an increased vacuolar volume. This increase in vacuolar size decreases the accessible cytoplasm, thereby increasing the mitochondrial volume fraction, and increases ATP3 mRNA localization to the mitochondria. While we believe the geometric constraints of the cell are important for mRNA localization, we cannot completely exclude the possibility that there are secondary factors such as receptor concentrations that also contribute to the mRNA localization, yet as we do not find them to be correlated with mRNA localization across all of our diverse perturbations, we do not believe receptor concentrations are the primary drivers of the mRNA localization changes we observe.

Our results also indicate that translation duration plays an important role in mRNA localization to the mitochondria. We believe that translation duration and mitochondrial volume fraction are interconnected, as both will impact the probability of the mitochondria interacting with a competent mRNA, meaning the MTS is exposed while still attached to the mRNA as a nascent-polypeptide (Figure 7A). ATP3 mRNA, which has fast translation elongation compared to TIM50, is in a competent state for a shorter period of time, so during fermentative conditions it is less likely to interact with mitochondria while in a competent state (Figure 7B). Conversely, ribosome stalling along TIM50 mRNA prolongs the time the mRNA is in a competent state, and this increases its probability of interacting with mitochondria even when in low mitochondrial volume fraction conditions (Figure 7B). As the mitochondrial volume fraction of cells increases as they shift to respiratory conditions, there is a higher probability of competent mRNAs interacting with the mitochondria, even if they are only competent for a short time (Figure 7B). While we show that one way to increase translation duration is a translation elongation stall caused by polyproline sequences, other mechanisms that increase translation duration including increasing ORF length, rare codons, and mRNA structures (Schuller and Green, 2018) have similar potential to impact mRNA localization.

Figure 7

Download asset Open asset

Mitochondrial volume fraction controls protein synthesis of nuclear-encoded mitochondrial proteins.

(A) Competency of mRNA localization to mitochondria is regulated by translational elongation. After the MTS is translated and exposed from the ribosome the mRNA becomes competent for localization to the mitochondria for as long as the mRNA is associated with the nascent chain. Slow elongation will extend the time an mRNA is competent. Mitochondrially localized mRNAs experience an increase in RNA stability and protein translatability. (B) Mitochondria can coordinate gene expression during times of metabolic need via mitochondrial-volume-fraction-based control and simple chemical kinetics of nuclear-encoded mRNA localization. *TIM50*-type: mRNAs with high affinity to mitochondria are always associated with mitochondria and thus not much affected by geometrical features. *ATP3*-type: mRNAs with low affinity for mitochondria localization are greatly affected by geometrical features of cells, that is mitochondrial volume fraction. Fast translation elongation leads to quick release of the mRNA/nascent chain complex, and results in a quick return to a non-competent state. When mitochondrial volume fraction is high in respiratory conditions, mRNA localization to mitochondria is increased, and protein synthesis is induced by its localization.

The localization of ribosomes on the yeast mitochondrial surface has been known since the 1970 s, implicating localized translation. It has been proposed that the functional relevance of this localized translation may be to facilitate co-translational import (Lesnik et al., 2015). The necessity of co-translational import is unclear, as in vitro mitochondrial import systems have found it to be important for the import of some proteins but not for others (Fujiki and Verner, 1991; Schatz, 1979; Suissa and Schatz, 1982). Our work points to localization to mitochondria having an alternative function: to upregulate protein synthesis, as we found that mRNA localization to mitochondria was sufficient to increase protein synthesis, whereas tethering mRNAs away from the mitochondria to the plasma membrane reduced protein synthesis. The decrease in protein levels when ATP3 mRNA is anchored to the plasma membrane is also associated with a growth defect during respiratory conditions. Yet localization to the mitochondria is not absolutely required for optimal growth; when ATP3 mRNAs were targeted away from the mitochondria to the ER, there was no growth defect. This may be because, similar to the mitochondria, there is an upregulation of protein levels when mRNAs are targeted to the ER.

Condition-dependent mRNA localization to the mitochondria as a means to control gene expression could be used to tune the protein composition of the mitochondria to the metabolic needs of the cell. Several observations support this hypothesis. For example, the conserved subunits of ATP synthase are all sensitive to the translation elongation rates in the cell, and these subunits showed similar localization regulation patterns (Table 1); this suggests that a mechanism may have evolved that coordinates the expression and stoichiometry of vital subunits of this complex. Interestingly it has been shown that nuclear-encoded components of the F₁ ATP synthase complex upregulate the translation of the mitochondrially encoded ATP6 and ATP8 subunits potentially contributing to the synchronization of these separated genomes (Rak and Tzagoloff, 2009). Further supporting the use of condition-dependent mRNA localization in impacting gene expression is global data showing that cycloheximide-sensitive mRNAs (ATP3-type) show a more than 2-fold increase in protein synthesis relative to constitutively localized mRNAs (TIM50-type) when cells shift from fermentation to respiration (Figure 5E). For ATP3-type mRNAs, the increase in protein synthesis during respiratory conditions is driven by both increases in translational efficiency and increased mRNA levels (Figure 1C–E, Figure 5—figure supplement 2). Similarly, the increase in protein levels seen when mRNAs were tethered to the mitochondria was driven by both an increase in mRNA levels and ribosome engagement (Figure 6C–E, Figure 6—figure supplement 1). As these reporter mRNAs are under identical promoters, we favor the hypothesis that localization to the mitochondria increases mRNA stability. It is unclear whether translational efficiency and mRNA stability are independently affected by localization to the mitochondria or whether they are interrelated, as previous reports have shown that increased translation initiation can lead to protection from mRNA degradation (Roy and Jacobson, 2013). As discussed previously, ribosome stalling increases mRNA binding competency, driving mRNA localization to the mitochondria, along with increased translation initiation and mRNA stability. As translation initiation is thought to be the rate-limiting step in translation for most mRNAs, this would explain our somewhat-contradictory results showing that more ribosome stalling, through polyproline insertion, leads to increased protein synthesis.

How mitochondrially localized mRNAs can increase protein synthesis is still an open question. We consider the simplest explanation, that there may be an increased density of ribosomes on the mitochondrial surface, as previously suggested in Drosophila oogenesis (Zhang et al., 2016). It also might be true that translation initiation factors or mRNA decay factors have altered activities near mitochondria that produce high levels of ATP. An alternative idea is that there are specialized ribosomes, enriched at the mitochondrial surface, that enhance translation (Xue and Barna, 2012). Intriguingly, it was recently found that specific ribosomal protein paralogs are necessary for normal mitochondrial function. Rpl1b∆ cells were found to be deficient for growth on nonfermentable carbon sources, whereas rpl1a∆ had no growth defect in these conditions. Furthermore, rpl1b∆ cells were found to have decreased translation of many mitochondrial proteins during respiratory conditions, including the ATP3-type mRNAs ATP1 and ATP2 (Segev and Gerst, 2018). Determining how mRNA localization increases protein synthesis will increase our understanding on how mitochondria are able to control their composition in relation to the metabolic needs of the cell.

While we have found this gene expression control mechanism in yeast, we speculate that higher eukaryotic cells could also use mitochondrial mRNA localization, controlled by translation elongation and mitochondrial volume fraction, to regulate protein synthesis. Hinting at this, a recent paper exploring global mRNA localization in mammalian cells found that many ATP synthase mRNAs had enhanced localization to the mitochondria in a cycloheximide-dependent manner, similar to ATP3-type mRNAs in yeast (Fazal et al., 2019). Further exploration into the role of mitochondrial volume fraction and translation elongation in mRNA localization to the mitochondria will provide insight into the regulation of mitochondria in health and disease.

Materials and methods

Yeast strains and plasmids

Request a detailed protocol

The yeast strains and plasmids used are listed in Supplementary file 1 and the oligonucleotides used for plasmid construction are described in Supplementary file 2. To reduce variability among the constructed yeast strains, the strains were created either through integration of a linear PCR product or a plasmid linearized through restriction digest. Yeast strains harboring the genomic MS2 sequence were constructed using pSH47 and pLOXHIS5MS2L, a gift from the J. Gerst laboratory, as previously described (Gadir et al., 2011; Haim-Vilmovsky and Gerst, 2009). To constitutively visualize mRNAs, pRS405CYC1p-MS2-4xGFP (TTP080) was constructed as follows: MCP-4xGFP sequence was PCR amplified from pMS2CPGFP(x4) (Gadir et al., 2011) and integrated into the pRS405 vector under the CYC1 promoter. The strains expressing MCP-4xGFP from 2 sets of integrated plasmids were selected through microscopy screening and used for further experiments. To visualize the mitochondria matrix, pRS406GPDp-Su9-mCherry was constructed as follows: Su9 (subunit 9 of the F₀ ATPase (Su9(1-69))) sequence was PCR amplified from pvt100-dsRed (Rafelski et al., 2012) and the yomCherry sequence was PCR amplified from pFA6a-link-yomCherry-SpHis5 (Lee et al., 2013), a gift from the K. Thorn laboratory, and integrated into pRS406 vector under a GPD promoter. The strains expressing Su9-mCherry from two sets of integrated plasmids were selected through microscopy screening. To construct plasmids shown in Figure 3A, (TTP134 and TTP136), endogenously MS2 sequence-tagged genes were amplified from genomic DNA of strains TTY1160 and TTY439, and inserted into pRS403 vector plasmids (TTP133 and TTP135), respectively. Annealed cytERM (C450_2C1) primers were integrated into TTP133 and TTP135 by combination of Gibson assembly and PCR. To construct TIM50-∆MTS plasmid (TTP245), MTS region 4–300 bp was swapped from TTP133 by combination of Gibson assembly and PCR. GFP tagging was performed with PCR mediated homologous recombination using pFA6a-link-yoGFP-SpHis5 (Lee et al., 2013) and integrations were confirmed by PCR. Deletion mutant strains were constructed with PCR mediated homologous recombination using pFA6a-hphMX6, a gift from the J. Wilhelm laboratory, and integrations were confirmed by PCR. To construct the plasmids used in Figure 2 and 3, yoGFP fragment was PCR-amplified from pFA6a-link-yoGFP-SpHis5 (Lee et al., 2013) or an iRFP fragment was PCR-amplified from NHB084 with a flag sequence and inserted into TTP133 to construct TTP155 and TTP146, respectively. Synthetic reporters were created by Gibson Assembly of a PCR-amplified TIM50 (1-300nt), TIM50 (301-1431nt), ATP3 (1-300nt) and/or ATP3 (301-936nt). To construct the polyproline deletion constructs TTP174 and TTP179, deletions were introduced by PCR-based site-directed mutagenesis. To endogenously insert MCP proteins into the C-terminus of Tom70p, Tom20p, Sec63p, we generated TTP119 and TTP153 by replacing a PCR amplified MCP fragment from pMS2CPGFP(x4) (Gadir et al., 2011) with yoGFP of pFA6a-link-yoGFP-SpHis5 and pFA6a-link-yoGFP-CaUra3 (Lee et al., 2013). MCP tagging was performed with PCR mediated homologous recombination using TTP119 and TTP153. Integrations were confirmed by PCR followed by sequencing. Plasma membrane anchor MCP-CaaX was constructed by PCR-based site-directed mutagenesis using TTP080. The strains expressing MCP-2xGFP-CaaX from two sets of integrated plasmids were selected through microscopy screening and used for further experiments.

Microscopy

Request a detailed protocol

Single molecule mRNA visualization with mitochondria was performed as follows: Yeast cells were grown in YPA medium containing 2% glucose (fermentative) or 3% glycerol + 2% ethanol (respiratory) with 15 ml glass tube at 30°C with rotator speeds of 60 rpm. 300 μl of mid-log phase wild-type yeast cells (OD600 of 0.4–0.7), grown in appropriate medium, were harvested and placed into an Y04C microfluidic chamber, controlled by the CellASIC Onix system. 300 μl YPAD were placed into the flow-wells, the chambers were loaded with cells at 3 psi, and medium was continuously flowed at 3 psi. Cells were imaged at 30°C with a Yokogawa CSUX Spinning Disk Confocal (Solamere Technology Group) mounted on a Nikon Eclipse Ti chassis motorized inverted microscope, located at the Department of Developmental and Cell Biology (UCI). Imaging was performed using a 100x/1.49 NA oil APO TIRF objective with the correction collar set manually for each experiment and a 1x tube lens (pixel size 0.084 μm). Z-stacks (300 nm steps) were acquired in the fluorescent channel (33 ms exposure) on a Hamamatsu electron-multiplying charge-coupled device (EMCCD) camera. Imaging was controlled using MicroManager ImageAcquisition (v1.4.16). For CHX treatment, a flow-well containing a final concentration of 100 μg/ml Cycloheximide (C7698; Sigma-Aldrich) was open and an image was taken every 3 s. To directly analyze the effect of bona fide inhibition of translation, we supplemented 1,10-Phenanthroline (131377; Sigma-Aldrich), which inhibits transcription, to CHX with final concentration of 250 µg/ml. For LTM treatment, a flow-well containing a final concentration of 50 μM Lactimidomycin (506291; MilliporeSigma) was open for 20 min and an image was taken every 3 s. Image data for Figure 2E and Figure 2—figure supplement 3 as well as C-terminus integrated GFP intensity data were collected as follows: Mid-log phase wild-type yeast cells (OD600 of 0.4 to 0.7) were grown in appropriate medium and 100 μl were placed into a 96-well Glass Bottom Plate (Cellvis LLC). For chloramphenicol treatment, cells were grown in YPAD medium containing a final concentration of 1 mg/ml chloramphenicol through the experiment (~16 hr) (C0378; Sigma-Aldrich). Cells were imaged at 23°C with 3 s interval by a Perkin Elmer UltraView Vox Spinning Disk Confocal mounted on an Olympus IX81 inverted microscope with Yokogawa CSU-X1-A1 spinning disk head, located at the UCSD School of Medicine Microscopy Core. Imaging was performed using either 100x/1.4 NA or 60x/1.42 NA oil objective with the correction collar set manually for each experiment (pixel size 0.068 µm or 0.111 μm). Z-stacks (300 nm steps) were acquired in the fluorescent channel on an EMCCD Hamamatsu 14 bit 1K × 1K camera. Imaging was controlled using Volocity (PerkinElmer). Image data for Figure 3—figure supplement 1 were collected as follows: Mid-log phase wild-type yeast cells (OD600 of 0.4 to 0.7) were grown in YPD medium and 100 μl were placed into a 96-well Glass Bottom Plate (Cellvis LLC). Cells were imaged at 23°C with 4 s interval by an Eclipse Ti2-E Spinning Disk Confocal with Yokogawa CSU-X1 (Yokogawa) with 50 µm pinholes, located at the Nikon Imaging Center UCSD. Imaging was performed using SR HP APO TIRF 100 × 1.49 NA oil objective with the correction collar set manually for each experiment (pixel size 0.074 µm). Z-stacks (300 nm steps) were acquired by a Prime 95B sCMOS camera (Photometrics). Imaging was controlled using NIS-Elements software (Nikon).

Reconstruction of 3D mitochondria and mRNA visualization

Request a detailed protocol

To allow accurate visualization of mRNA molecules, multiple MS2 stem-loops are inserted in the 3’-UTR of the mRNA of interest and are recognized by the MCP-GFP fusion protein (Bertrand et al., 1998; Haim-Vilmovsky and Gerst, 2009). We improved this system by titrating down the MCP-GFP levels until we observed single molecule mRNA foci, which we verified by single molecule RNA FISH (smFISH) (Hocine et al., 2013; Tutucci et al., 2018a; Figure 1—figure supplement 1). We then performed rapid 3D live cell imaging using spinning disk confocal microscopy. We reconstructed and analyzed the spatial relationship between the mRNAs and mitochondria using custom ImageJ plugin Trackmate (Tinevez et al., 2017) and MitoGraph V2.0, which we previously developed to reconstruct 3D mitochondria based on matrix marker fluorescent protein intensity (Rafelski et al., 2012; Viana et al., 2015; Figure 1A). We measured the distance between mRNA and mitochondria by finding the closest meshed surface area of the mitochondria matrix. Bias-reduced logistic regression (Firth, 1992; Firth, 1993) was used to determine which factors influenced the manual tracking of foci in Trackmate. Signal-to-noise ratio (SNR), median intensity of foci, and minimum distance between tracked foci were screened for their contributions to manual tracking of foci. Two-sided p-values were compared for the different variables. The logistic regression analysis shows that median intensity (p-value=0.046) and SNR (p-value=0.036) are the two features detectable by human eyes to track the foci. The code used for the analysis is available from https://github.com/tsuboitat (Tsuboi and Viana, 2020; copy archived at https://github.com/elifesciences-publications/Mitograph_Distance).

Validation of single molecule mRNA of MS2-MCP system

Request a detailed protocol

It has been observed that version 4 of the MS2-MCP system is prone to cytoplasmic aggregation, thought to be caused by either stabilization of mRNA or its degradation intermediates by MCP proteins (Tutucci et al., 2018a). We improved the MS2-MCP system by titrating down the MCP-GFP levels until we could observe single molecule mRNA foci, which we verified by single molecule RNA FISH (smFISH) (Hocine et al., 2013; Figure 1—figure supplement 1). Analysing the co-localization rate of candidate mRNA ORF and MS2 sequences using anti-sense probes (Figure 1—figure supplement 1A), we confirmed that more than 84% of the ORFs and MS2 sequences are a continuous stretch in the same mRNA (Figure 1—figure supplement 1B). The number of MCP-GFP foci in live cells is statistically no different (Wilcoxon rank-sum test, p-value<0.05, N > 43) than the number of smFISH foci in the tested mRNAs (Figure 1—figure supplement 1C). We concluded that the MCP-GFP foci in the live cell images are reliable single molecule mRNAs and not degradation intermediates.

Single-Molecule FISH and image acquisition and analysis

Request a detailed protocol

Single-molecule FISH (smFISH) was performed as previously described (Tutucci et al., 2018b). Yeast strains were grown at 23°C in YPAD medium containing 2% glucose and harvested at OD600 0.6–0.8. Cells were fixed by adding paraformaldehyde (15714; 32% solution, EM grade; Electron Microscopy Science) to a final concentration of 4%. Cells walls were removed by resuspension in spheroplast buffer (buffer B containing 20 mM VRC (S1402S; Ribonucleoside–vanadyl complex NEB), and 25U of Lyticase enzyme (L2524; Sigma). Digested cells were seeded on 18 mm polylysine-treated coverslips. The hybridization mix (10% formamide (205821000; ACROS organics), 2x SSC, 1 mg/ml BSA, 10 mM VRC, 5 mM NaHPO4 pH 7.5, 1 mg/μL E. coli tRNA, 1 mg/µl ssDNA) was prepared with probe mix (final concentration 125 nM). Cells were then hybridized at 37°C for 4 hr in the dark. Nuclei were stained with 0.5 μg/mL DAPI in 1x PBS. Coverslips were mounted on glass slides using ProLong Gold Antifade Mountant (ThermoFisher). Images were acquired using an Olympus BX61 wide-field epifluorescence microscope with a 100x/1.35 NA UPlanApo objective. Samples were visualized using an X-Cite 120 PC lamp (EXFO) and the ORCA-R2 Digital CCD camera (Hamamatsu). Metamorph software (Molecular Devices) was used for acquisition. Z-sections were acquired at 200 nm intervals over an optical range of 8.0 μm. Image pixel size: XY, 64.5 nm. smFISH probes for TIM50, ATP3 and TOM22 probes were designed using and purchased from the Stellaris Probe Designer by LGC Biosearch Technologies. MS2 sequence probes, gifted from the R. Singer laboratory, were synthetized by Invitrogen-Thermo Fisher and labelled in the lab using Cy3 dyes (Amersham) as previously described (Tutucci et al., 2018b). Quantification of the FISH spots in the cytoplasm was performed using a custom ImageJ pipeline and manually detected. If the focus in different channels was within two pixels, it was considered to be co-localized; otherwise, it was considered to be two non-overlapping and distinct foci.

Definition of ‘localization’ and ‘association’ to mitochondria

Request a detailed protocol

To analyze the association of mRNA to mitochondria, we first defined the ‘localized’ threshold as twice the size of the mode (190 nm) of the distance between TIM50 mRNA and mitochondria, which were treated with translation elongation and transcription inhibitors cycloheximide (CHX) and 1,10-Phenanthroline (PHE), respectively, expecting that the majority of TIM50 mRNAs associate with the mitochondrial surface in these conditions (Figure 1—figure supplement 2). We further classified mRNA localization to reflect a stable or transient association using this 0.19 um threshold. Since the translational elongation rate is 9.5aa/sec (Shah et al., 2013), which implies that it takes more than 30 s to translate reporter genes (more than 300aa), we defined a mitochondria-associated mRNA as being co-localized to the mitochondria for at least 3 s (two consecutive time points).

Segmentation of cell boundaries and analysis of cell volume

Request a detailed protocol

We segmented cell boundaries from the GFP field of the image, which we acquired for mRNA localization analysis in Figure 1B, using the ImageJ plug-in Trainable Weka Segmentation (Arganda-Carreras et al., 2017). Cell volume was calculated through this boundary. To analyze the effective mitochondrial volume fraction for analyzing mutant strains, we segmented the non-cytoplasmic area as vacuole. To acquire the volume fraction of the vacuole, we chose the center focal plane of each cell in bright field image and segmented the outer cell shape and inner black area, which we assumed as non-cytoplasmic area. Then, fit the ellipse to both outer cell shape and inner non-cytoplasmic area and calculate the volume using shorter radios of the fitted ellipse as depth.

Single molecule mRNA in silico experiment and mathematical modeling

Request a detailed protocol

We used Paraview (http://www.paraview.org) to create 3D meshes for cell and mitochondria boundaries based on the segmentations. To model ideal Brownian particles in the cytoplasm, we generated particle trajectories using a random walk (diffusion coefficients (D) = 0.1 µm²/s) (Wu et al., 2016) in the space between the cell wall and mitochondria surface for at least 40,000 trajectories per cell. Then the proportion of mitochondrial localization per cell (as we defined previously) was quantified. Because the particles were generated as space filling between the cell wall and mitochondrial surface and the mitochondria is a tubular structure, the proportion of mitochondrial localization R_localized was linearly correlated with mitochondria volume fraction as described in the following equation (Figure 2—figure supplement 2A, black line).

R_{l o c a l i z e d} \propto \frac{m i t o c h o n d r i a l s u r f a c e a r e a}{c e l l v o l u m e} \propto m i t o c h o n d r i a l v o l u m e f r a c t i o n

We assessed whether the localization behavior for TOM22 mRNA, which we assumed to be a non-mitochondrial localized mRNA, matched with that of a Brownian particle. Although the proportion of mitochondrial localization for TOM22 mRNA was linearly correlated with the mitochondrial volume fraction, the slope was different from that of a Brownian particle. We hypothesized this was because we had not considered the other cell compartments, such as nucleus (7% of cell volume) (Jorgensen et al., 2007), vacuoles (10% of cell volume) (Chan et al., 2016), and mitochondria itself (10% of cell volume; this study). We determined that the difference in this proportion was caused by this organelle volume and to account for it, we minimized the cell size accordingly. To fit the TOM22 mRNA localization proportion line to the Brownian particles we set the effective cell volume to 66%. This was below our expectation, but we found this acceptable since we had not included other organelles such as ER or lipid droplets in the virtual cell to study localization.

We assumed that mitochondrial localized mRNA concentration [mRNA_localized], free diffusing mRNA concentration [mRNA_free], and the concentration of mitochondrial surface point where mRNA can bind [mito_surface] are in thermodynamical equilibrium. We then hypothesized that some mRNAs may associate with the mitochondrial surface by a sequence-specific association (e.g. via an MTS) with a constant K such that:

[m R N A_{f r e e}] [{m i t o}_{s u r f a c e}] ⇄ [m R N A_{l o c a l i z e d}]

K = \frac{[{m R N A}_{l o c a l i z e d}]}{[{m i t o}_{s u r f a c e}] [{m R N A}_{f r e e}]}

$K$ is also thermodynamically given by

K = e^{\frac{- Δ G_{l o c a l i z a t i o n}}{k_{B} T}}

where k_B is Boltzmann’s constant, T is temperature, and $Δ G_{l o c a l i z a t i o n}$ is the free energy gain from mitochondrial localization. For the Brownian particle-like TOM22, we assumed a $Δ G_{l o c a l i z a t i o n}$ value of 0; therefore, TOM22 had the lowest equilibrium constant, which we called K₀. We then hypothesized that any arbitrary mRNA has an equilibrium constant of K = AK₀, where A thus represents a fold-change in affinity between that mRNA and mitochondria compared to a freely diffusing mRNA. As the value of A increased in the simulation, the proportion of mitochondrial localization of the mRNA for a given mitochondrial volume fraction increased as well (Figure 2—figure supplement 2B). The proportion of mitochondrial localization $R_{l o c a l i z e d}'$ for an arbitrary mRNA when K = AK₀ is fitted to the following equation using the linear trend line for proportion of localization $R_{l o c a l i z e d_{0}}$ of TOM22 mRNA. Upon the same mitochondrial volume fraction, we acquire $R_{l o c a l i z e d}'$ as

R_{l o c a l i z e d}' = A ∙ K_{0} [{m i t o}_{s u r f a c e}] [1 - R_{l o c a l i z e d}']

R_{l o c a l i z e d}^{'} = A \frac{R_{l o c a l i z e d_{0}}}{[1 - R_{l o c a l i z e d_{0}}]} [1 - R_{l o c a l i z e d}^{'}]

R_{l o c a l i z e d}^{'} = \frac{A \cdot R_{l o c a l i z e d_{0}}}{1 + (A - 1) R_{l o c a l i z e d_{0}}}

Thus, we show that the probability of an mRNA localizing to a mitochondrion is dependent upon both the mitochondrial concentration and the free energy change of localizing.

Quantification of protein expression in each cell from image data

Request a detailed protocol

Quantification of GFP-tagged mitochondrially-localized protein expression was performed using a custom analysis pipeline. The MitoGraph was applied to the mCherry channel to identify the locations of mitochondria and the resulting binary mask was then segmented to determine projections of individual mitochondria. Next, the sum of the intensity of the GFP channel covered by binary mask from each cell was measured using a custom ImageJ script.

Western blotting

Request a detailed protocol

Yeast cells were grown in YPA medium containing 2% glucose or 3% glycerol + 2% ethanol. The cells were harvested when the culture reached an OD600 of 0.8. The protein products of various GFP reporter genes were detected by western blotting using an anti-GFP antibody (11814460001; Roche) and a Mouse IgG (H+L) Secondary Antibody (32430; Thermo Fisher). For the detection of the flag-tagged proteins, anti-flag antibody (F1804; Sigma-Aldrich) was used. A mouse anti-a-tubulin antibody (12G10, Developmental Studies Hybridoma Bank) was used as a loading control. The intensities of the bands on the blots were quantified using Molecular Imager ChemiDoc XRS+ System (Bio-Rad) and ImageJ. The relative levels of the protein products were determined by comparison with a standard curve prepared using a series of dilutions.

Polysome analysis

Request a detailed protocol

Yeast cells were grown exponentially at 30°C and harvested by centrifugation. Cell extracts were prepared with mortar and pestle as described previously (Tsuboi et al., 2012). The equivalent of 10 A260 units was then layered onto linear 10% to 50% (w/w) sucrose density gradients. Sucrose gradients (10%–50% sucrose in 20 mM HEPES [pH 7.4], 5 mM magnesium chloride, 100 mM potassium chloride, 2 mM DTT, 100 µg/ml CHX) were prepared in 14 × 89 mm polyallomer tubes (Beckman Coulter) with a gradient master. Crude extracts were layered on top of the sucrose gradients and were centrifuged at 30,000 rpm in a SW-28 rotor for 3 hr at 4°C. Gradients were fractionated using a gradient fractionator and UA-6 detector (ISCO/BRANDEL). RNA from ribosome-free and ribosome-bound fraction (monosome – polysome) as shown in Figure 6—figure supplement 2 were purified from each fraction using guanidine hydrochloride and processed for RT-qPCR.

RT-qPCR

Request a detailed protocol

RNA was extracted using the MasterPure Yeast RNA Purification Kit (Epicentre). cDNA was prepared using ProtoScript II Reverse Transcriptase (NEB #M0368X) with a combination of oligo(dT) primers and random hexamers according to the manufacturer’s instructions. mRNA abundance was determined by qPCR using SYBR Green PCR Master Mix (Applied Biosystems) and primers specific for FLAG-GFP (ATGGACTACAAGGACGACG and CCTCACCACTCACAGAAAAC), OM14 (TCACCACCACAACAATAAGAAG and AGCGTCAAAAGACCCAGAG), OM45 (ACAGTAATCCTTTGAAACGCC and TCACCCCATCCTTCTAAACC), and ACT1 (GAGAGGCGAGTTTGGTTTCA and TCACCCGGCCTCTATTTTC). The mRNA levels were normalized to ACT1 abundance, and the fold change between samples was calculated by a standard ∆∆Ct analysis.

Data and materials availability

Request a detailed protocol

The code used for analyzing the distance between mRNA and mitochondrial surface and generating random walk trajectory is available from https://github.com/tsuboitat. Further information and requests for resources, scripts, and reagents should be directed to and will be fulfilled by the lead contact, T.T. (ttsuboi@ucsd.edu).

Data availability

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

The following previously published data sets were used

(2016) NCBI Gene Expression Omnibus
ID GSE74454. Synchronized translation programs across cellular compartments.

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE74454
(2014) NCBI Gene Expression Omnibus
ID GSE61011. Targeting and Plasticity of Mitochondrial Proteins Revealed by Proximity-Specific Ribosome Profiling.

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE61011
1. Morgenstern M
2. Stiller SB
3. Lübbert P
4. Peikert CD
5. Dannenmaier S
6. Drepper F
7. Weill U
8. Höß P
9. Feuerstein R
10. Gebert M
11. Bohnert M
12. van der Laan M
13. Schuldiner M
14. Schütze C
15. Oeljeklaus S
16. Pfanner N
17. Wiedemann N
18. Warscheid B
(2017) ProteomeXchange
ID PXD006151. Definition of a high-confidence mitochondrial proteome at quantitative scale_pure versus crude mitochondria.

http://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD006151

References

(2017) Autophagy induction under carbon starvation conditions is negatively regulated by carbon catabolite repression
Journal of Biological Chemistry 292:19905–19918.
https://doi.org/10.1074/jbc.M117.817510
- PubMed
- Google Scholar
(2017) Trainable weka segmentation: A machine learning tool for microscopy pixel classification
Bioinformatics 33:2424–2426.
https://doi.org/10.1093/bioinformatics/btx180
- PubMed
- Google Scholar
1. Attardi G
2. Schatz G
(1988) Biogenesis of mitochondria
Annual Review of Cell Biology 4:289–331.
https://doi.org/10.1146/annurev.cb.04.110188.001445
- PubMed
- Google Scholar
1. Bertrand E
2. Chartrand P
3. Schaefer M
4. Shenoy SM
5. Singer RH
6. Long RM
(1998) Localization of ASH1 mRNA particles in living yeast
Molecular Cell 2:437–445.
https://doi.org/10.1016/S1097-2765(00)80143-4
- PubMed
- Google Scholar
1. Borst P
2. Grivell LA
(1978) The mitochondrial genome of yeast
Cell 15:705–723.
https://doi.org/10.1016/0092-8674(78)90257-X
- Google Scholar
1. Brody S
(1992) Circadian rhythms in Neurospora crassa: the role of mitochondria
Chronobiology International 9:222–230.
https://doi.org/10.3109/07420529209064531
- PubMed
- Google Scholar
(2003) The S. cerevisiae HAP complex, a key regulator of mitochondrial function, coordinates nuclear and mitochondrial gene expression
Comparative and Functional Genomics 4:37–46.
https://doi.org/10.1002/cfg.254
- PubMed
- Google Scholar
1. Chan YH
2. Reyes L
3. Sohail SM
4. Tran NK
5. Marshall WF
(2016) Organelle size scaling of the budding yeast vacuole by relative growth and inheritance
Current Biology 26:1221–1228.
https://doi.org/10.1016/j.cub.2016.03.020
- PubMed
- Google Scholar
(2016) Synchronized mitochondrial and cytosolic translation programs
Nature 533:499–503.
https://doi.org/10.1038/nature18015
- Google Scholar
1. De Deken RH
(1966) The crabtree effect: a regulatory system in yeast
Journal of General Microbiology 44:149–156.
https://doi.org/10.1099/00221287-44-2-149
- PubMed
- Google Scholar
1. Egner A
2. Jakobs S
3. Hell SW
(2002) Fast 100-nm resolution three-dimensional microscope reveals structural plasticity of mitochondria in live yeast
PNAS 99:3370–3375.
https://doi.org/10.1073/pnas.052545099
- Google Scholar
1. Eliyahu E
2. Pnueli L
3. Melamed D
4. Scherrer T
5. Gerber AP
6. Pines O
7. Rapaport D
8. Arava Y
(2010) Tom20 mediates localization of mRNAs to mitochondria in a translation-dependent manner
Molecular and Cellular Biology 30:284–294.
https://doi.org/10.1128/MCB.00651-09
- PubMed
- Google Scholar
(2009) The mitochondrial proteome database: mitop2
Methods in Enzymology 457:3–20.
https://doi.org/10.1016/S0076-6879(09)05001-0
- PubMed
- Google Scholar
1. Fazal FM
2. Han S
3. Parker KR
4. Kaewsapsak P
5. Xu J
6. Boettiger AN
7. Chang HY
8. Ting AY
(2019) Atlas of subcellular RNA localization revealed by APEX-Seq
Cell 178:473–490.
https://doi.org/10.1016/j.cell.2019.05.027
- PubMed
- Google Scholar
Book
1. Firth D
(1992) Bias reduction, the Jeffreys prior and GLIM
In: Fahrmeir L, Francis B, Gilchrist R, Tutz G, editors. Advances in GLIM and Statistical Modelling. New York: Springer. pp. 91–100.
https://doi.org/10.1007/978-1-4612-2952-0
- Google Scholar
1. Firth D
(1993) Bias reduction of maximum likelihood estimates
Biometrika 80:27–38.
https://doi.org/10.1093/biomet/80.1.27
- Google Scholar
1. Fujiki M
2. Verner K
(1991)
Coupling of protein synthesis and mitochondrial import in a homologous yeast in vitro system

The Journal of Biological Chemistry 266:6841–6847.
- PubMed
- Google Scholar
(2011) Localization of mRNAs coding for mitochondrial proteins in the yeast Saccharomyces cerevisiae
RNA 17:1551–1565.
https://doi.org/10.1261/rna.2621111
- Google Scholar
1. Garcia M
2. Darzacq X
3. Delaveau T
4. Jourdren L
5. Singer RH
6. Jacq C
(2007) Mitochondria-associated yeast mRNAs and the biogenesis of molecular complexes
Molecular Biology of the Cell 18:362–368.
https://doi.org/10.1091/mbc.e06-09-0827
- PubMed
- Google Scholar
1. Garcia M
2. Delaveau T
3. Goussard S
4. Jacq C
(2010) Mitochondrial presequence and open reading frame mediate asymmetric localization of messenger RNA
EMBO reports 11:285–291.
https://doi.org/10.1038/embor.2010.17
- Google Scholar
(1975) Effect of chloramphenicol on replication of mitochondria in Tetrahymena
Journal of Cellular Physiology 85:59–72.
https://doi.org/10.1002/jcp.1040850108
- PubMed
- Google Scholar
1. Golani-Armon A
2. Arava Y
(2016) Localization of nuclear-encoded mRNAs to mitochondria outer surface
Biochemistry 81:1038–1043.
https://doi.org/10.1134/S0006297916100023
- Google Scholar
(2017) Visualization of cytosolic ribosomes on the surface of mitochondria by electron cryo-tomography
EMBO Reports 18:1786–1800.
https://doi.org/10.15252/embr.201744261
- PubMed
- Google Scholar
1. Haim-Vilmovsky L
2. Gerst JE
(2009) m-TAG: a PCR-based genomic integration method to visualize the localization of specific endogenous mRNAs in vivo in yeast
Nature Protocols 4:1274–1284.
https://doi.org/10.1038/nprot.2009.115
- Google Scholar
1. Hocine S
2. Raymond P
3. Zenklusen D
4. Chao JA
5. Singer RH
(2013) Single-molecule analysis of gene expression using two-color RNA labeling in live yeast
Nature Methods 10:119–121.
https://doi.org/10.1038/nmeth.2305
- Google Scholar
1. Hübscher V
2. Mudholkar K
3. Chiabudini M
4. Fitzke E
5. Wölfle T
6. Pfeifer D
7. Drepper F
8. Warscheid B
9. Rospert S
(2016) The Hsp70 homolog Ssb and the 14-3-3 protein Bmh1 jointly regulate transcription of glucose repressed genes in Saccharomyces cerevisiae
Nucleic Acids Research 44:5629–5645.
https://doi.org/10.1093/nar/gkw168
- Google Scholar
(2002) Systematic identification of pathways that couple cell growth and division in yeast
Science 297:395–400.
https://doi.org/10.1126/science.1070850
- PubMed
- Google Scholar
(2007) The size of the nucleus increases as yeast cells grow
Molecular Biology of the Cell 18:3523–3532.
https://doi.org/10.1091/mbc.e06-10-0973
- PubMed
- Google Scholar
(1974)
Cytoplasmic type 80 s ribosomes associated with yeast mitochondria. ii. evidence for the association of cytoplasmic ribosomes with the outer mitochondrial membrane in situ

The Journal of Biological Chemistry 249:3297–3303.
- PubMed
- Google Scholar
1. Lee S
2. Lim WA
3. Thorn KS
(2013) Improved blue, green, and red fluorescent protein tagging vectors for S. cerevisiae
PLOS ONE 8:e67902.
https://doi.org/10.1371/journal.pone.0067902
- PubMed
- Google Scholar
(2014) OM14 is a mitochondrial receptor for cytosolic ribosomes that supports co-translational import into mitochondria
Nature Communications 5:1–10.
https://doi.org/10.1038/ncomms6711
- Google Scholar
(2015) Localized translation near the mitochondrial outer membrane: An update
RNA Biology 12:801–809.
https://doi.org/10.1080/15476286.2015.1058686
- Google Scholar
1. Marc P
2. Margeot A
3. Devaux F
4. Blugeon C
5. Corral-Debrinski M
6. Jacq C
(2002) Genome-wide analysis of mRNAs targeted to yeast mitochondria
EMBO Reports 3:159–164.
https://doi.org/10.1093/embo-reports/kvf025
- PubMed
- Google Scholar
1. Martin KC
2. Ephrussi A
(2009) mRNA localization: gene expression in the spatial dimension
Cell 136:719–730.
https://doi.org/10.1016/j.cell.2009.01.044
- PubMed
- Google Scholar
1. Morgenstern M
2. Stiller SB
3. Lübbert P
4. Peikert CD
5. Dannenmaier S
6. Drepper F
7. Weill U
8. Höß P
9. Feuerstein R
10. Gebert M
11. Bohnert M
12. van der Laan M
13. Schuldiner M
14. Schütze C
15. Oeljeklaus S
16. Pfanner N
17. Wiedemann N
18. Warscheid B
(2017) Definition of a High-Confidence mitochondrial proteome at quantitative scale
Cell Reports 19:2836–2852.
https://doi.org/10.1016/j.celrep.2017.06.014
- PubMed
- Google Scholar
1. Paulo JA
2. O'Connell JD
3. Everley RA
4. O'Brien J
5. Gygi MA
6. Gygi SP
(2016) Quantitative mass spectrometry-based multiplexing compares the abundance of 5000 S. cerevisiae proteins across 10 carbon sources
Journal of Proteomics 148:85–93.
https://doi.org/10.1016/j.jprot.2016.07.005
- Google Scholar
(2014) Local slowdown of translation by nonoptimal codons promotes nascent-chain recognition by SRP in vivo
Nature Structural & Molecular Biology 21:1100–1105.
https://doi.org/10.1038/nsmb.2919
- PubMed
- Google Scholar
1. Rafelski SM
2. Viana MP
3. Zhang Y
4. Chan YH
5. Thorn KS
6. Yam P
7. Fung JC
8. Li H
9. Costa LF
10. Marshall WF
(2012) Mitochondrial network size scaling in budding yeast
Science 338:822–824.
https://doi.org/10.1126/science.1225720
- PubMed
- Google Scholar
1. Rak M
2. Tzagoloff A
(2009) F1-dependent translation of mitochondrially encoded Atp6p and Atp8p subunits of yeast ATP synthase
PNAS 106:18509–18514.
https://doi.org/10.1073/pnas.0910351106
- Google Scholar
1. Reid DW
2. Nicchitta CV
(2015) Diversity and selectivity in mRNA translation on the endoplasmic reticulum
Nature Reviews Molecular Cell Biology 16:221–231.
https://doi.org/10.1038/nrm3958
- PubMed
- Google Scholar
1. Roy B
2. Jacobson A
(2013) The intimate relationships of mRNA decay and translation
Trends in Genetics 29:691–699.
https://doi.org/10.1016/j.tig.2013.09.002
- PubMed
- Google Scholar
1. Rühle T
2. Leister D
(2015) Assembly of F1F0-ATP synthases
Biochimica Et Biophysica Acta (BBA) - Bioenergetics 1847:849–860.
https://doi.org/10.1016/j.bbabio.2015.02.005
- Google Scholar
1. Saint-Georges Y
2. Garcia M
3. Delaveau T
4. Jourdren L
5. Le Crom S
6. Lemoine S
7. Tanty V
8. Devaux F
9. Jacq C
(2008) Yeast mitochondrial biogenesis: a role for the PUF RNA-binding protein Puf3p in mRNA localization
PLOS ONE 3:e2293.
https://doi.org/10.1371/journal.pone.0002293
- PubMed
- Google Scholar
1. Schatz G
(1979) How mitochondria import proteins from the cytoplasm
FEBS Letters 103:201–211.
https://doi.org/10.1016/0014-5793(79)81328-9
- Google Scholar
1. Schuller AP
2. Green R
(2018) Roadblocks and resolutions in eukaryotic translation
Nature Reviews Molecular Cell Biology 19:526–541.
https://doi.org/10.1038/s41580-018-0011-4
- Google Scholar
1. Segev N
2. Gerst JE
(2018) Specialized ribosomes and specific ribosomal protein paralogs control translation of mitochondrial proteins
Journal of Cell Biology 217:117–126.
https://doi.org/10.1083/jcb.201706059
- Google Scholar
1. Shah P
2. Ding Y
3. Niemczyk M
4. Kudla G
5. Plotkin JB
(2013) Rate-limiting steps in yeast protein translation
Cell 153:1589–1601.
https://doi.org/10.1016/j.cell.2013.05.049
- PubMed
- Google Scholar
1. Suissa M
2. Schatz G
(1982)
Import of proteins into mitochondria

The Journal of Biological Chemistry 257:13048–13055.
- Google Scholar
(2017) TrackMate: An open and extensible platform for single-particle tracking
Methods 115:80–90.
https://doi.org/10.1016/j.ymeth.2016.09.016
- Google Scholar
1. Tsuboi T
2. Kuroha K
3. Kudo K
4. Makino S
5. Inoue E
6. Kashima I
7. Inada T
(2012) Dom34:Hbs1 Plays a General Role in Quality-Control Systems by Dissociation of a Stalled Ribosome at the 3′ End of Aberrant mRNA
Molecular Cell 46:518–529.
https://doi.org/10.1016/j.molcel.2012.03.013
- Google Scholar
Software
1. Tsuboi T
2. Viana MP
(2020) Mitograph_Distance, version fed4102
GitHub.

https://github.com/tsuboitat/Mitograph_Distance
1. Tutucci E
2. Vera M
3. Biswas J
4. Garcia J
5. Parker R
6. Singer RH
(2018a) An improved MS2 system for accurate reporting of the mRNA life cycle
Nature Methods 15:81–89.
https://doi.org/10.1038/nmeth.4502
- Google Scholar
(2018b) Single-mRNA detection in living S. cerevisiae using a re-engineered MS2 system
Nature Protocols 13:2268–2296.
https://doi.org/10.1038/s41596-018-0037-2
- Google Scholar
(1993) Kinetics of growth and sugar consumption in yeasts
Antonie van Leeuwenhoek 63:343–352.
https://doi.org/10.1007/BF00871229
- Google Scholar
Book
(2015) Quantifying Mitochondrial Content in Living Cells
Elsevier.
https://doi.org/10.1016/bs.mcb.2014.10.003
- Google Scholar
(2014) Targeting and plasticity of mitochondrial proteins revealed by proximity-specific ribosome profiling
Science 346:748–751.
https://doi.org/10.1126/science.1257522
- Google Scholar
(1971) Induction of the cytoplasmic petite mutation in Saccharomyces cerevisiae by the antibacterial antibiotics erythromycin and chloramphenicol
Molecular and General Genetics MGG 111:209–223.
https://doi.org/10.1007/BF00433106
- Google Scholar
1. Wu B
2. Eliscovich C
3. Yoon YJ
4. Singer RH
(2016) Translation dynamics of single mRNAs in live cells and neurons
Science 352:1430–1435.
https://doi.org/10.1126/science.aaf1084
- Google Scholar
1. Xue S
2. Barna M
(2012) Specialized ribosomes: a new frontier in gene regulation and organismal biology
Nature Reviews Molecular Cell Biology 13:355–369.
https://doi.org/10.1038/nrm3359
- Google Scholar
1. Yan X
2. Hoek TA
3. Vale RD
4. Tanenbaum ME
(2016) Dynamics of translation of single mRNA molecules in Vivo
Cell 165:976–989.
https://doi.org/10.1016/j.cell.2016.04.034
- PubMed
- Google Scholar
1. Zhang Y
2. Chen Y
3. Gucek M
4. Xu H
(2016) The mitochondrial outer membrane protein MDI promotes local protein synthesis and mt DNA replication
The EMBO Journal 35:1045–1057.
https://doi.org/10.15252/embj.201592994
- Google Scholar
1. Zhang D
2. Shan SO
(2012) Translation elongation regulates substrate selection by the signal recognition particle
Journal of Biological Chemistry 287:7652–7660.
https://doi.org/10.1074/jbc.M111.325001
- PubMed
- Google Scholar
1. Zid BM
2. O’Shea EK
(2014) Promoter sequences direct cytoplasmic localization and translation of mRNAs during starvation in yeast
Nature 514:117–121.
https://doi.org/10.1038/nature13578
- Google Scholar

Decision letter

Alan G Hinnebusch

Reviewing Editor; Eunice Kennedy Shriver National Institute of Child Health and Human Development, United States
James L Manley

Senior Editor; Columbia University, United States

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

Acceptance summary:

Your paper is of great interest in providing evidence that localization of yeast mRNAs to mitochondria increases their translational efficiencies, and that the increased localization of certain mRNAs during a shift from fermentative to respiratory growth is driven by two factors: (i) the increased proportion of cell volume made up by mitochondria in respiratory growth, and (ii) for certain mRNAs including TIM50, a stretch of consecutive proline codons, known to confer a pause during translation elongation, which they hypothesize is required to increase the time available for association of the mitochondrial targeting sequence in the nascent polypeptide with its receptor on the mitochondrial outer membrane.

Decision letter after peer review:

Thank you for submitting your article "Mitochondrial volume fraction and translation speed impact mRNA localization to the mitochondria and protein synthesis" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by James Manley as the Senior Editor. The reviewers have opted to remain anonymous.

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

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). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, as they do with your paper, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Our expectation is that the authors will eventually carry out the additional experiments and report on how they affect the relevant conclusions either in a preprint on bioRxiv or medRxiv, or if appropriate, as a Research Advance in eLife, either of which would be linked to the original paper.

Summary:

This paper provides evidence that localization of yeast mRNAs to mitochondria (mito) increases their translational efficiencies, and that the increased mito-localization seen for certain nucleus-encoded mito protein mRNAs during a shift from fermentative (ferm) to respiratory (resp) growth is driven by two factors. One is the increased proportion of cell volume made up by mito in respiratory growth, while the second, at least for TIM50 mRNA under ferm conditions when the mito volume fraction is low, is a stretch of consecutive proline codons, known to confer a pause during translation elongation, which they hypothesize is required to increase the time available for association of the mito targeting sequence (MTS) in the nascent polypeptide with the MTS receptor on the mito outer membrane and thereby enhance mRNA localization to mito-associated ribosomes. ATP3 appears to lack an elongation-stalling sequence as inserting the polyproline tract into its mRNA can enhance its mito-association in ferm conditions, which is normally lower than that seen for TIM50 that naturally contains the PolyPro. This model can also explain the previous observation that inhibiting translation elongation with CHX can enhance mito-association of ATP3 and certain other mito mRNAs, but not of TIM50, presumably by substituting for translation elongation stalling sequences in those former mRNAs. Consistent with their model, inhibiting initiation vs. elongation with lactimidomycin did not enhance mito. localization as this would prevent synthesis of the MTS. They also did tethering experiments and showed that artificially tethering any reporter mRNA tested (even GFP) to mitochondria increases its translation and mRNA abundance, whereas tethering TIM50 or ATP3 , but not TOM22 or GFP, to the plasma membrane (PM) reduces their translation relative to the untethered state.

Revisions for this paper:

The following are the major comments that would likely require additional experimental evidence to be addressed in a satisfactory way:

1) New experiments are needed to measure the levels of the TOM complex, or other relevant complexes (OM14 or SAM-Mdm10), to see if they increase under resp. growth conditions, which if so, could be an important factor beyond mito. volume per se in increasing localization of mito. mRNAs.

2) Present additional imaging data (videos) to convince us that the same mRNAs are being examined in different z-stacks.

3) Provide evidence for increased translation initiation, beyond just showing increased mRNA localization, in response to manipulations that increase the mitochondria volume fraction.

4) Show that deleting the MTS, rather than replacing it with an ER-targeting sequence, impairs mito. localization.

5) Add analysis of control constructs to confirm that the sequence encoding Polyproline increases mito. localization owing to translation elongation through polyproline codons rather than an inhibitory effect of the RNA sequences per se, involving synonomous codon replacements encoding PolyPro or a single base pair insertion to place the polyPro tract out of frame.

6) Provide a more comprehensive analysis of mRNA localization/translation in both fermentative and resp. conditions.

In addition, the paper should be revised, possibly including the title, to acknowledge that the present evidence is inadequate to conclude that mitochondrial volume is a major driver of mito protein localization, and to note that other intrinsic features of the translated ORFs (e.g. MTS, codon usage, ORF length), changes in translation and the level of the constituents involved in tethering under respiratory conditions, and alterations in cellular architecture (e.g. organelle size), may be equally or more important. The paper should also be revised to address as many as possible of the other major comments not touched on above

Reviewer #1:

This paper is interesting in providing evidence that localization of yeast mRNAs to mitochondria (mito) increases their translational efficiencies, and that the increased mito-localization seen for certain nucleus-encoded mito protein mRNAs during a shift from fermentative (ferm) to respiratory (resp) growth is driven by two factors. One is the increased proportion of cell volume made up by mito in respiratory growth, and although this point is established only by correlation, they did a thorough job of modulating the mito volume proportion in different ways and found coherent effects on mito localization of mRNAs. The second important element for mito localization, at least for TIM50 mRNA under ferm conditions when the mito volume fraction is low, appears to be a stretch of consecutive proline codons, known to confer a pause during translation elongation, which they hypothesize is required to increase the time available for association of the mito targeting sequence (MTS) in the nascent polypeptide with the MTS receptor on the mito outer membrane and thereby enhance mRNA localization to mito-associated ribosomes. ATP3 appears to lack an elongation-stalling sequence as inserting the polyproline tract into its mRNA can enhance its mito-association in ferm conditions, which is normally lower than that seen for TIM50 that naturally contains the PolyPro. This model can also explain the previous observation that inhibiting translation elongation with CHX can enhance mito-association of ATP3 and certain other mito mRNAs, but not of TIM50, presumably by substituting for translation elongation stalling sequences in those former mRNAs. Consistent with their model, inhibiting initiation vs. elongation with lactimidomycin did not enhance mito. localization as this would prevent synthesis of the MTS. They also did tethering experiments and showed that artificially tethering any reporter mRNA tested (even GFP) to mitochondria increases its translation and mRNA abundance, whereas tethering TIM50 or ATP3 , but not TOM22 or GFP, to the plasma membrane (PM) reduces their translation relative to the untethered state.

As listed below, there are a number of important experiments omitted and controls lacking. They have failed to establish clearly that the MTS of TIM50 is required for mito association of the mRNA, showing only that substituting it with an ER-localization sequence reduces the mito localization. They also need to show that simply deleting the MTS impairs mito localization of TIM50 mRNA, if this has not already been published, in order to firmly establish that the MTS is actually necessary for mito localization of TIM50. Additional control experiments are needed regarding the polyproline tract to distinguish between an effect of the mRNA sequence on localization independently of translation, versus the putative effect of Pro codons in stalling elongation: rather than just deleting it from TIM50, they should also have made synonymous codon replacements by mutating the wobble position of each Pro codon. And in addition to just inserting it in frame in ATP3, they should have made a single base, out-of-frame insertion in which the in-frame codons would no longer encode PolyPro. Finally, they should show that the PolyPro tract won't work if it is located too close to the MTS, as the elongation pause needs to be enacted only after the complete MTS has emerged from the ribosome exit tunnel.

There were also a number of cases where they did not examine the effects of manipulating TIM50 or ATP3 under resp conditions in addition to ferm conditions (or at least I couldn't tell if both conditions were examined) making it unclear whether the elongation pause is needed for efficient localization only under ferm growth when the mito cell volume is relatively low,

– Ratio of association is a poor identifier of measurements, such as in Figure 1B, as ratio of what to what is not defined. If it's mitochondrial-associated to total mRNA, proportion would be a better term. Also, the language in figure legends can sometimes be misconstrued to indicate a ratio of association between fermentative and respiratory growth; and so more precision is needed in all of the figures where this term is used.

– As noted above, the authors failed to show that the MTS of TIM50 is actually necessary for mito localization, only that its replacement with an ER-targeting signal reduces mito association. They need a deletion of the MTS versus replacement to make the point convincingly.

– Figure 3D: The legend claims that both ferm and resp conditions were examined, but it appears from the text that these data are only in ferm conditions, unless "ratio" of association means something different than above. In any event, the data in both conditions should be provided.

– Figure 4C-E: the same difficulty as for Figure 3D exists here, of unclear labeling and vague descriptions in the legends. Results from both ferm and resp conditions are needed to determine if the effects of polyproline on localization are restricted to only ferm conditions when mito volume is low.

– The experiments on the polyproline stretch need controls of synonymous codon changes vs. deletion for TIM50; and insertion of Pro codons out of frame for ATP3, to distinguish effects of mRNA sequence vs elongation stalling on localization.

– In the tethering experiments of Figure 6, one would like to know whether the effect on tethering of ATP3 is less in resp vs. ferm conditions; and it's not clear which condition was actually analyzed. They also need to add the control of measuring native ATP3 and TIM50 expression in the strain with Tom70-MCP, as their expression should be unaltered.

– It's unclear why PM-tethering reduces translation of mRNAs that are normally mito-localized, but not non-localized mRNAs, as the findings with TOM22 and GFP mRNAs with MS2 insertions show clearly that there are abundant ribosomes near the PM for translation of these mRNAs, and all of the mRNAs targeted to mito show increased translation. Presumably there are other unknown factors involved allowing TIM50 and ATP3 to be translated better when they are tethered to mitochondria vs the PM, but the authors don't consider this issue and should comment on it.

Reviewer #2:

The gene expression control events highlighted by this work are interesting and speak to the finely tuned nature of a cell to respond to cellular and environmental change. The data is of high quality and well presented. I also find the data to be thought provoking, but I am not convinced that mRNA localization is directly linked to mitochondrial volume fraction vs. binding interactions driven by the nascent chain and difference in translation across metabolic programs. Overall, I am supportive of publication if the points raised below can be adequately addressed.

1) How does mitochondrial composition change with organelle growth? It is not addressed how the TOM complex, or other relevant complexes (OM14 or SAM-Mdm10), change with mitochondrial volume or respiratory activity. For instance, if the density of these complexes are maintained as the membrane grows, or even increase with respiration, this would mean an increase in the total number of binding sites on the mitochondrial surface. As import of MTS proteins and insertion of Tom22 involves one or more of these complexes, it may be changes in receptor numbers that drive association of mRNAs through the nascent polypeptide chain, not volume per se.

2) Is ORF length a driver of mitochondrial localization? Of the mRNAs studied, the proteins encoded are 476 (TIM50), 311 (ATP3), and 152 (TOM22) residues in length. Yet, the authors do not address the correlation of length with the percent mRNA localized. This question is furthered by the observation that among the ATP complex subunits, the shortest ORF (ATP16, 160 residues) is the one that is not impacted by CHX. Williams et al. 2014 has noted that there was correlation between ORF length and localization, with the majority of nascent chains observed at mitochondria being >180 residues in length. A potentially informative experiment to address this issue would involve assaying mRNA localization and protein expression levels in TIM50/ATP3/TOM22-GPF reporters in response to fermentative vs. respiratory growth conditions.

3) The authors perform mRNA imaging over a time series and define a mRNA as localized if it is within a defined distance of the mitochondria in two consecutive imaging frames ~3 seconds apart. Mitochondria can move, an mRNA could release and rebind between z-stacks, or different mRNAs could bind sites physically close together at different times. As such, how is it that the authors know they are counting the same mRNA in consecutive z-stacks? This should be discussed and example images/videos provided to detail how this issue is dealt with. Assuming there is not much change between time points, the authors should be able to provide binding durations for each mRNA. I would expect such dwell time information to be valuable in terms of the modeling data in Figure 2 and could be used to favor or possibly exclude models explaining how/why different mRNAs are localized to the mitochondria surface (e.g. co-translational import, engagement with mitochondrial associated ribosomes, or transient interactions with translational/RNA-binding proteins needed for translation).

Reviewer #3:

In this manuscript, Tsuboi et al. investigated the mRNAs localization of nuclear encoded mitochondrial proteins in yeast under fermentative and respiratory growth conditions. They used single mRNA imaging with MS2 system in live cells with fluorescently labeled mitochondria. The authors observed that three types of behaviors: constitutively (TIM50), partially (ATP3) and not localized (TOM22). They found that ATP3 localization to mitochondria increases when yeast was changed from fermentative to respiratory condition, while TIM50 and TOM22 have less changes. The mRNA localization was correlated to mitochondrial volume fraction. The localization depends on the MTS downstream coding sequence and the authors claim that the localization increased translation initiation. By mutation and translation inhibition, the authors found out that the decreased translation elongation localizes mRNA to mitochondria. Finally, by comparing mis-localized RNAs, the authors found out decreased protein production and some growth defects. The authors proposed a model that mitochondrial RNA localization increases protein translation.

Here the authors used single molecule approach to measure the distance between mRNA and the mitochondria directly, which is technically impressive and nice. Overall, the reviewer think that the data basically agrees with the previous transcriptome wide RNA localization data and the model proposed there: translation drives RNA localization on to the mitochondria, the degree of localization and the CHX sensitivity depend on the relative position of MTS and the downstream sequences in the mRNA. However, I think the data and the proposed model has some gaps, that requires tuning down the claim or further experimental confirmation.

1) The authors concentrated on MTS driven mitochondrial mRNA localization. As proposed before, it mainly depends on the location of MTS within the ORF and the downstream sequence (the length and the plausible stalling sequence). The authors observed that the association with mitochondria of certain mRNAs (ATP3) increased when yeast shifted from fermentative to respiratory condition. The authors claim that mitochondria volume fraction "regulate" the fraction of association. I think this is a chicken and egg question. The translation efficiencies of mRNAs may also explain these observations. ATP3 mRNAs are not actively translated in fermentation condition. The translation efficiency dramatically increases (translation regulation) when the growth condition changed to respiration, which in turn drives the localization of the mRNAs to mitochondria. It makes sense that when fraction of mitochondria is increased, there is a higher chance for MTS to meet the receptor on it, which results in higher colocalization. The authors used ways to manipulate mitochondria fraction and indeed observed more interaction. But in those experiment, there is no direct evidence that increased localization results more translation initiation.

2) Mathematical modeling: the authors claim that ATP3 and TIM50 mRNAs has certain affinity to mitochondria. The authors claim the affinity to be "strength of mRNA-specific association". But if the localization is mainly due to translation and MTS binding to the receptor, what does these binding affinities mean? Should it just be the function of translation efficiencies of mRNAs? Should the translation of mRNA explain all the localization? If the model can differentiate whether localization of mRNA increases translation efficiency, it would be more convincing and exciting.

3) The authors claimed increased translation initiation based on western, GFP fluorescence. Those final protein products may also depend on the protein stability. For example, if TIM50 or ATP3 are not synthesized on the surface of mitochondria, the nascent proteins may have lower stability and get degraded, because they may require co-transaltional translocation. The decreased protein can also result in growth phenotype.

4) The authors used combination of endogenous labeling and reporters. Sometimes, it is confusing which strains were used because it was not clearly stated. For example, MCP-GFP and ATP3-flag-GFP cannot be in the same strain, right? First, it cannot be used to image RNA. Second, the qPCR measurement of mRNA level using GFP primer include both MCP-GFP and ATP3-flag-GFP (Figure 6). So are there MCP-RFP used to visualize RNAs?

https://doi.org/10.7554/eLife.57814.sa1

Author response

Revisions for this paper:

The following are the major comments that would likely require additional experimental evidence to be addressed in a satisfactory way:

1) New experiments are needed to measure the levels of the TOM complex, or other relevant complexes (OM14 or SAM-Mdm10), to see if they increase under resp. growth conditions, which if so, could be an important factor beyond mito. volume per se in increasing localization of mito. mRNAs.

Firstly, we want to emphasize that the mitochondrial volume fraction, not mitochondrial volume is important for mRNA localization in terms of space restriction/ cellular constraints as shown for TOM22 mRNA localization and simulation results (Figure 2B, 2D, Figure 2—figure supplement 2). From our modeling we are able to recapitulate general trends in our experimental data by using mRNAs with constant affinities across conditions and just adjusting the spatial restrictions from increasing mitochondrial volume fraction. While our simulation doesn’t completely recapitulate our experimental data, and this could be partially explained by changes in relevant receptors, we think it is beyond the scope of this paper to completely characterize all aspects of mRNA localization in this single manuscript. To touch on this issue though we explored as suggested the expression of the TOM complex, and other relevant complexes (OM14 or SAM-Mdm10) in fermentative versus respiratory conditions using data from Morgenstern et al., 2017. This dataset shows a 3.5 fold increase in ATP3 and no change in TIM50 when they compare protein expression levels in glucose and glycerol which are very similar to our results (Figure 1D, E). From their experiments, they showed protein expression are changed as follows.

The protein expression of TOM and most other complexes did not change appreciably in glucose and glycerol, except for OM14 and OM45. As we have perturbations that change mitochondrial volume fraction and ATP3 localization independent of nutrients we sought to explore if OM14 and OM45 change in these perturbations. We were able to recapitulate the large changes in OM14 and OM45 RNA levels by RT-qPCR in respiratory conditions, but we see a slight decrease of OM14 and no significant change in OM45 in sch9∆ and reg1∆ strains, where we see a significant increase in mitochondrial volume fraction and ATP3 mRNA localization. This doesn’t mean OM14 and OM45 play no role in localization during respiratory conditions but that mitochondrial volume fraction is a key factor impacting mRNA localization. We believe further exploration of this relationship is beyond the scope of this manuscript.

We have added the following to the Results section:

“An alternative possibility is that expression of the import machinery or the NAC receptor correlate with mitochondrial volume fraction and that these are what is important for the changes in ATP3 mRNA localization. […] While we are able to recapitulate the large increase in expression of OM14 and OM45 in respiratory conditions, we do not find a significant increase in the expression of these RNAs in reg1∆ and sch9∆ mutant strains (Figure 2—figure supplement 5).”

And to the Discussion section: “While we believe the geometric constraints of the cell are important for mRNA localization, we cannot completely exclude the possibility that there are secondary factors such as receptor concentrations that also contribute to the mRNA localization, yet as we do not find them to be correlated with mRNA localization across all of our diverse perturbations, we do not believe receptor concentrations are the primary drivers of the mRNA localization changes we observe.”

2) Present additional imaging data (videos) to convince us that the same mRNAs are being examined in different z-stacks.

We included videos for TIM50, ATP3 and TOM22 mRNAs in fermentative conditions as Video 2. We also added videos for TIM50 and TOM22 mRNAs with addition of cycloheximide and 1,10-phenanthroline for Figure 1—figure supplement 2 as Video 1.

3) Provide evidence for increased translation initiation, beyond just showing increased mRNA localization, in response to manipulations that increase the mitochondria volume fraction.

We show that mRNAs that are elongation rate sensitive in their localization to the mitochondria have a small (~15%) but significant increase in translation efficiency upon a switch to respiratory conditions when these mRNAs are found to be localized to the mitochondria (Figure 5—figure supplement 2A). For the specific mRNAs (ATP2 and ATP3) that we have directly shown have increased mRNA localization in respiratory versus fermentative conditions, we see a 32% and 42% increase in translation efficiency respectively during respiratory conditions (Source Data 1).

Overall we have toned back the language on the increase in translation initiation, and focused instead on the increase in protein synthesis that accompanies mRNAs being localized to the mitochondria as we feel that both mRNA stability and translation efficiency changes are important for the fine tuning of gene expression driven by mRNA localization.

4) Show that deleting the MTS, rather than replacing it with an ER-targeting sequence, impairs mito. localization.

Because of Covid we did not have access to the original microscopes, so instead we used another microscope and visualized mRNA with 4 second interval. Cells were imaged at 23℃ with an Eclipse Ti2-E Spinning Disk Confocal with Yokogawa CSU-X1 (Yokogawa) with 50 µm pinholes, located at the Nikon Imaging Center UCSD. Imaging was performed using SR HP APO TIRF 100x 1.49 NA oil objective with the correction collar set manually for each experiment (pixel size 0.074µm). Z-stacks (300nm steps) were acquired by a Prime 95B sCMOS camera (Photometrics). Imaging was controlled using NIS-Elements software. We have added this as Figure 3—figure supplement 1.

5) Add analysis of control constructs to confirm that the sequence encoding Polyproline increases mito. localization owing to translation elongation through polyproline codons rather than an inhibitory effect of the RNA sequences per se, involving synonomous codon replacements encoding PolyPro or a single base pair insertion to place the polyPro tract out of frame.

Our lactimidomycin data (Figure 3A) shows that translation initiation is key for the localization of TIM50 mRNA to the mitochondria, and that upon lactimidomycin treatment TIM50 and ATP3 have indistinguishable localization phenotypes even though TIM50 mRNAs contain the 7 consecutive polyprolines and ATP3 doesn’t, arguing against a role for the RNA sequence directly having a role in the differential localization between TIM50 and ATP3.

6) Provide a more comprehensive analysis of mRNA localization/translation in both fermentative and resp. conditions.

We added newly analyzed data of proportion of association for the chimeric reporters and TIM50-P7∆ reporter genes in respiratory conditions. These data are included as Figure 3—figure supplement 2 and Figure 4D.

In addition, the paper should be revised, possibly including the title, to acknowledge that the present evidence is inadequate to conclude that mitochondrial volume is a major driver of mito protein localization, and to note that other intrinsic features of the translated ORFs (e.g. MTS, codon usage, ORF length), changes in translation and the level of the constituents involved in tethering under respiratory conditions, and alterations in cellular architecture (e.g. organelle size), may be equally or more important. The paper should also be revised to address as many as possible of the other major comments not touched on above

We have changed the title to reflect the more inclusive language of translation duration, as we agree that other factors that impact translation duration can impact localization. We have added the following paragraph to the Discussion “While we show that one way to increase translation duration is a translation elongation stall caused by polyproline sequences, other mechanisms that increase translation duration including increasing ORF length, rare codons, and mRNA structures(Schuller and Green, 2018) have similar potential to impact mRNA localization.”

Reviewer #1:

[…]

There were also a number of cases where they did not examine the effects of manipulating TIM50 or ATP3 under resp conditions in addition to ferm conditions (or at least I couldn't tell if both conditions were examined) making it unclear whether the elongation pause is needed for efficient localization only under ferm growth when the mito cell volume is relatively low,

– Ratio of association is a poor identifier of measurements, such as in Figure 1B, as ratio of what to what is not defined. If it's mitochondrial-associated to total mRNA, proportion would be a better term. Also, the language in figure legends can sometimes be misconstrued to indicate a ratio of association between fermentative and respiratory growth; and so more precision is needed in all of the figures where this term is used.

Thank you for your clarification. We changed the y axis of Figures 1-5 and figure supplements to “proportion of associated RNA”. We also edited the figure legends of them.

– As noted above, the authors failed to show that the MTS of TIM50 is actually necessary for mito localization, only that its replacement with an ER-targeting signal reduces mito association. They need a deletion of the MTS versus replacement to make the point convincingly.

As discussed in major additional experiment 4 above.

– Figure 3D: The legend claims that both ferm and resp conditions were examined, but it appears from the text that these data are only in ferm conditions, unless "ratio" of association means something different than above. In any event, the data in both conditions should be provided.

As discussed in major additional experiment 6 above.

– Figure 4C-E: the same difficulty as for Figure 3D exists here, of unclear labeling and vague descriptions in the legends. Results from both ferm and resp conditions are needed to determine if the effects of polyproline on localization are restricted to only ferm conditions when mito volume is low.

As discussed in major additional experiment 6 above.

– The experiments on the polyproline stretch need controls of synonymous codon changes vs. deletion for TIM50; and insertion of Pro codons out of frame for ATP3, to distinguish effects of mRNA sequence vs elongation stalling on localization.

As discussed in major additional experiment 5 above.

– In the tethering experiments of Figure 6, one would like to know whether the effect on tethering of ATP3 is less in resp vs. ferm conditions; and it's not clear which condition was actually analyzed. They also need to add the control of measuring native ATP3 and TIM50 expression in the strain with Tom70-MCP, as their expression should be unaltered.

As it is written in the legend, we showed the protein expression level in respiratory conditions to see if altered protein expression leads slow growth in respiratory condition. For TIM50 and ATP3 we observed similar protein expression differences in fermentative/vegetative conditions, while for ATP2 there was less of an effect in fermentative conditions. As we were focused on the respiratory conditions where we see the slow growth effect we only included the impact of tethering away on protein production during those conditions.

Author response image 1

Download asset Open asset

Tethering to plasmamembrane reduce protein production in vegetative and fermentative condition.

N=3.

– It's unclear why PM-tethering reduces translation of mRNAs that are normally mito-localized, but not non-localized mRNAs, as the findings with TOM22 and GFP mRNAs with MS2 insertions show clearly that there are abundant ribosomes near the PM for translation of these mRNAs, and all of the mRNAs targeted to mito show increased translation. Presumably there are other unknown factors involved allowing TIM50 and ATP3 to be translated better when they are tethered to mitochondria vs the PM, but the authors don't consider this issue and should comment on it.

We believe this is not a TIM50 or ATP3 specific issue, and that any mRNA that is localized to the mitochondrial surface has increased protein synthesis as seen for the cytosolic protein Flag-GFP in Figure 6B, C. So it is not that the plasma membrane is specifically limiting protein synthesis of mito-mRNAs localized there, but that mRNAs pulled away from the mitochondria don’t get the beneficial effects of being localized to the mitochondria and the associated increased protein synthesis. Similarly when we decrease mRNA localization to the mitochondria by deleting prolines in TIM50, we see decreased protein synthesis, not because it is localized to the plasma membrane, but because it is not localized to the mitochondrial surface.

We inserted the following statement in Discussion.

“How mitochondrially localized mRNAs can increase protein synthesis is still an open question. […] Determining how mRNA localization increases protein synthesis will increase our understanding on how mitochondria are able to control their composition in relation to the metabolic needs of the cell.”

“Finally, they should show that the PolyPro tract won't work if it is located too close to the MTS, as the elongation pause needs to be enacted only after the complete MTS has emerged from the ribosome exit tunnel.“

We agree with this assumption that MTS needs to be exposed from the ribosome for binding to mitochondria. We inserted a polyproline tract at 100aa from MTS and still see the increased mRNA localization. The ribosome tunnel size is generally thought as 30aa length and we are uncertain if effects from inserting the polyproline tract closer than 100aat would be from the MTS not being exposed or because the MTS was disrupted as the MTS is assumed to be ~30-100aa, but not always perfectly defined.

Reviewer #2:

The gene expression control events highlighted by this work are interesting and speak to the finely tuned nature of a cell to respond to cellular and environmental change. The data is of high quality and well presented. I also find the data to be thought provoking, but I am not convinced that mRNA localization is directly linked to mitochondrial volume fraction vs. binding interactions driven by the nascent chain and difference in translation across metabolic programs. Overall, I am supportive of publication if the points raised below can be adequately addressed.

1) How does mitochondrial composition change with organelle growth? It is not addressed how the TOM complex, or other relevant complexes (OM14 or SAM-Mdm10), change with mitochondrial volume or respiratory activity. For instance, if the density of these complexes are maintained as the membrane grows, or even increase with respiration, this would mean an increase in the total number of binding sites on the mitochondrial surface. As import of MTS proteins and insertion of Tom22 involves one or more of these complexes, it may be changes in receptor numbers that drive association of mRNAs through the nascent polypeptide chain, not volume per se.

This is fair point and we assessed change of receptors. Please see major additional experiment 1 above.

2) Is ORF length a driver of mitochondrial localization? Of the mRNAs studied, the proteins encoded are 476 (TIM50), 311 (ATP3), and 152 (TOM22) residues in length. Yet, the authors do not address the correlation of length with the percent mRNA localized. This question is furthered by the observation that among the ATP complex subunits, the shortest ORF (ATP16, 160 residues) is the one that is not impacted by CHX. Williams et al. 2014 has noted that there was correlation between ORF length and localization, with the majority of nascent chains observed at mitochondria being >180 residues in length. A potentially informative experiment to address this issue would involve assaying mRNA localization and protein expression levels in TIM50/ATP3/TOM22-GPF reporters in response to fermentative vs. respiratory growth conditions.

We agree with the reviewers points that length of the mRNA will also potentially impact localization to the mitochondrial surface and have changed the title to “translation duration” to better encompass this point. Yet we have also found that length is not the only factor as we can remove the 7 polyprolines from TIM50 (a <2% length reduction) and see a large drop in localization, while inserting these short residues in ATP3 increases localization. We also believe that using fluorescent microscopy to explore the impact of length skew the results.

We tested TIM50-iRFP, ATP3-iRFP by inserting iRFP into C-terminus of those genes. Please see major additional experiment 6 above. We see differences in ratio of mitochondrial association with/without iRFP. We observed ratio of association as 0.52 in fermentative and 0.7 in respiratory condition for TIM50 mRNA and ratio of association as 0.15 in fermentative and 0.5 in respiratory condition for ATP3 mRNA. When we insert iRFP, we observed an increase in the ratio of association for ATP3-iRFP (0.27 in fermentative and 0.6 in respiratory condition) yet we see a slight decrease in ratio of association for TIM50-iRFP (0.37 in fermentative and 0.64 in respiratory condition) compared to the shorter TIM50. The additional iRFP length of 948nt which can be at least 10-20nm in vivo (Adivarahan et al., 2018) may affect our quantification of localization because of the thresholding we set for a localized mRNA.

3) The authors perform mRNA imaging over a time series and define a mRNA as localized if it is within a defined distance of the mitochondria in two consecutive imaging frames ~3 seconds apart. Mitochondria can move, an mRNA could release and rebind between z-stacks, or different mRNAs could bind sites physically close together at different times. As such, how is it that the authors know they are counting the same mRNA in consecutive z-stacks? This should be discussed and example images/videos provided to detail how this issue is dealt with. Assuming there is not much change between time points, the authors should be able to provide binding durations for each mRNA. I would expect such dwell time information to be valuable in terms of the modeling data in Figure 2 and could be used to favor or possibly exclude models explaining how/why different mRNAs are localized to the mitochondria surface (e.g. co-translational import, engagement with mitochondrial associated ribosomes, or transient interactions with translational/RNA-binding proteins needed for translation).

Visualizing single molecule mRNA and mitochondria simultaneously in 3D was technically challenging experiment. MS2-GFP signals from mRNAs are fast bleaching and we can visualize them for 30 sec, which we are not able to analyze the binding duration time. As described in Materials and methods, we tracked mRNA manually and tested whether those tracks are equal to nearest neighbor search. We cannot exclude the opportunity that we mistrack the same single molecule mRNA in consecutive time points. As we added in major additional experiment 2 above, you may easily track the single mRNA, which associated to mitochondria by eyes, but for freely diffusing mRNAs we rely on justification by nearest neighbor search. We focused on the analysis of ratio of mitochondrial associated mRNA, which binds to mitochondria for consecutive 2 time points, which we are much more confident on those being the same mRNAs.

Reviewer #3:

In this manuscript, Tsuboi et al. investigated the mRNAs localization of nuclear encoded mitochondrial proteins in yeast under fermentative and respiratory growth conditions. They used single mRNA imaging with MS2 system in live cells with fluorescently labeled mitochondria. The authors observed that three types of behaviors: constitutively (TIM50), partially (ATP3) and not localized (TOM22). They found that ATP3 localization to mitochondria increases when yeast was changed from fermentative to respiratory condition, while TIM50 and TOM22 have less changes. The mRNA localization was correlated to mitochondrial volume fraction. The localization depends on the MTS downstream coding sequence and the authors claim that the localization increased translation initiation. By mutation and translation inhibition, the authors found out that the decreased translation elongation localizes mRNA to mitochondria. Finally, by comparing mis-localized RNAs, the authors found out decreased protein production and some growth defects. The authors proposed a model that mitochondrial RNA localization increases protein translation.

Here the authors used single molecule approach to measure the distance between mRNA and the mitochondria directly, which is technically impressive and nice. Overall, the reviewer think that the data basically agrees with the previous transcriptome wide RNA localization data and the model proposed there: translation drives RNA localization on to the mitochondria, the degree of localization and the CHX sensitivity depend on the relative position of MTS and the downstream sequences in the mRNA. However, I think the data and the proposed model has some gaps, that requires tuning down the claim or further experimental confirmation.

1) The authors concentrated on MTS driven mitochondrial mRNA localization. As proposed before, it mainly depends on the location of MTS within the ORF and the downstream sequence (the length and the plausible stalling sequence). The authors observed that the association with mitochondria of certain mRNAs (ATP3) increased when yeast shifted from fermentative to respiratory condition. The authors claim that mitochondria volume fraction "regulate" the fraction of association. I think this is a chicken and egg question. The translation efficiencies of mRNAs may also explain these observations. ATP3 mRNAs are not actively translated in fermentation condition. The translation efficiency dramatically increases (translation regulation) when the growth condition changed to respiration, which in turn drives the localization of the mRNAs to mitochondria. It makes sense that when fraction of mitochondria is increased, there is a higher chance for MTS to meet the receptor on it, which results in higher colocalization. The authors used ways to manipulate mitochondria fraction and indeed observed more interaction. But in those experiment, there is no direct evidence that increased localization results more translation initiation.

We agree with the reviewer’s chicken and egg proposition, to decouple condition specific translation effects from mRNA localization effects we tested the impact of tethering reporter mRNAs to mitochondria during fermentative conditions. For all mRNAs tested we measured increased protein synthesis. We believe this is in part due to increased mRNA stability, but that doesn’t explain the full extent of increased protein production. So to directly assay whether localization increases ribosome loading we performed polysome profiling and found that there is an increased fraction of mRNA associated with ribosomes when an mRNA is localized to the mitochondria, this suggests that in part mitochondrial localization increases protein synthesis through increased translation initiation (Figure 6A-E).

2) Mathematical modeling: the authors claim that ATP3 and TIM50 mRNAs has certain affinity to mitochondria. The authors claim the affinity to be "strength of mRNA-specific association". But if the localization is mainly due to translation and MTS binding to the receptor, what does these binding affinities mean? Should it just be the function of translation efficiencies of mRNAs? Should the translation of mRNA explain all the localization? If the model can differentiate whether localization of mRNA increases translation efficiency, it would be more convincing and exciting.

As much of the actual localization process and import is still an active area of research with many unknowns we sought to use as simplistic model as possible to explore whether space/mitochondrial volume fraction with constant “affinity” could explain the condition specific localization phenotypes we see. From further research in our paper we agree that for ATP3 and TIM50 the time an mRNA is competent (MTS exposed) is probably the major factor impacting “strength of mRNA-specific association”. This competency will be impacted by multiple factors including translation initiation rate (translation efficiency), but also mRNA length and as we show elongation rate.

3) The authors claimed increased translation initiation based on western, GFP fluorescence. Those final protein products may also depend on the protein stability. For example, if TIM50 or ATP3 are not synthesized on the surface of mitochondria, the nascent proteins may have lower stability and get degraded, because they may require co-transaltional translocation. The decreased protein can also result in growth phenotype.

This unique idea that mis-co-translational translocation leads to protein degradation is a very interesting question and we were also interested in how the protein stability can change. We observed that half-life (Protein half-life was analyzed through western blotting. Samples were harvested at 0, 30 and 60 min after the addition of CHX (final 100µg/ml). The half-life was calculated by fitting data in log scale.) of the ATP3-GFP protein which is translated at the plasma membrane and the control half-life was both more than 2 hours, and not significantly different from each other. This is intuitive that 2 hours protein half-life does not much affect protein abundance of yeast cells, in which the cell cycle is 1.5 hours, but gene expression (cell cycle) controls protein abundance. Baum et al., 2019 has mathematically shown that cell cycle is more important than protein half-life for protein abundance when the protein half-life is longer than cell cycle. They stated that “It becomes evident that the influence of cell division increases with increasing half-life of either mRNA (Figure 3A, top row) or mRNA and protein.”

4) The authors used combination of endogenous labeling and reporters. Sometimes, it is confusing which strains were used because it was not clearly stated. For example, MCP-GFP and ATP3-flag-GFP cannot be in the same strain, right? First, it cannot be used to image RNA. Second, the qPCR measurement of mRNA level using GFP primer include both MCP-GFP and ATP3-flag-GFP (Figure 6). So are there MCP-RFP used to visualize RNAs?

For visualizing mRNA movements in Figure 3 and 4, we used iRFP as a reporter genes. For Figure 6 we used MCP-GFP as a control in comparison to mito- and ER- tethering and tested protein levels using GFP-antibody and conduct qPCR using flag sequence as a part of amplicon. GFP-antibody binds to both MCP-GFP and flagGFP reporter gene, however we can distinguish the difference by the size.

https://doi.org/10.7554/eLife.57814.sa2

Article and author information

Author details

Tatsuhisa Tsuboi
1. Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States
2. Department of Developmental and Cell Biology and Center for Complex Biological Systems, University of California Irvine, Irvine, United States
3. Division of Biological Science, Graduate School of Science, Nagoya University, Nagoya, Japan
Contribution
Conceptualization, Resources, Data curation, Software, Formal analysis, Fundingacquisition, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing -review and editing

For correspondence
ttsuboi@ucsd.edu

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-3249-030X
Matheus P Viana

Department of Developmental and Cell Biology and Center for Complex Biological Systems, University of California Irvine, Irvine, United States

Present address
Allen Institute for Cell Science, Seattle, United States

Contribution
Software, Methodology, Writing - review and editing

Competing interests
No competing interests declared
Fan Xu

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Formal analysis, Investigation

Competing interests
No competing interests declared
Jingwen Yu

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Software, Formal analysis, Investigation, Methodology

Competing interests
No competing interests declared
Raghav Chanchani

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Data curation, Validation, Methodology

Competing interests
No competing interests declared
Ximena G Arceo

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Writing - original draft

Competing interests
No competing interests declared
Evelina Tutucci

Department of Anatomy and Structural Biology, Albert Einstein College of Medicine, Bronx, United States

Present address
Systems Biology Lab, Amsterdam Institute of Molecular and Life Sciences, Vrije Universiteit Amsterdam, Amsterdam, Netherlands

Contribution
Methodology

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-1998-7146
Joonhyuk Choi

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Methodology

Competing interests
No competing interests declared
Yang S Chen

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Formal analysis, Methodology

Competing interests
No competing interests declared
Robert H Singer
1. Department of Anatomy and Structural Biology, Albert Einstein College of Medicine, Bronx, United States
2. Gruss-Lipper Biophotonics Center, Albert Einstein College of Medicine, Bronx, United States
3. Department of Neuroscience, Albert Einstein College of Medicine, Bronx, United States
4. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United States
Contribution
Resources, Funding acquisition, Methodology

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-6725-0093
Susanne M Rafelski

Department of Developmental and Cell Biology and Center for Complex Biological Systems, University of California Irvine, Irvine, United States

Present address
Allen Institute for Cell Science, Seattle, United States

Contribution
Conceptualization, Resources, Software, Supervision, Funding acquisition, Methodology, Project administration, Writing - review and editing

For correspondence
susanner@alleninstitute.org

Competing interests
No competing interests declared
Brian M Zid

Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, United States

Contribution
Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Investigation, Writing - original draft, Writing - review and editing, Project administration

For correspondence
zid@ucsd.edu

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-1876-2479

Funding

Japan Society for the Promotion of Science (Research abroad fellowship)

Tatsuhisa Tsuboi

Japan Society for the Promotion of Science (18J00995)

Tatsuhisa Tsuboi

Uehara Memorial Foundation (Research abroad fellowship)

Tatsuhisa Tsuboi

National Institute of General Medical Sciences (R35GM128798)

Brian M Zid

National Science Foundation (MCB-1330451)

Susanne M Rafelski

Ellison Medical Foundation (New Scholar Award In Aging)

Susanne M Rafelski

National Institute of General Medical Sciences (GM57071)

Robert H Singer

Swiss National Science Foundation (Fellowship P2GEP3_155692)

Evelina Tutucci

Swiss National Science Foundation (Fellowship P300PA_164717)

Evelina Tutucci

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

Acknowledgements

We thank members of the Zid laboratory as well as T Endo, S Iwasaki, G Goshima, E Koslover, W Wang and V Bilanchone for helpful discussions and A Subramaniam and S Mukherji for feedback on the paper. Receipt of the MS2-tagging plasmids and yeast-optimized fluorophore plasmids from Dr. J Gerst and Dr. K Thorn is gratefully acknowledged. We thank M Zid, A Guzikowski and V Harjono for critically reading the manuscript. This work was supported in part by National Institutes of Health GM57071 (to RHS), startup funds from UCI, NSF grant MCB-1330451 and Ellison Medical Foundation (to SMR), startup funds from UCSD and from the National Institutes of Health R35GM128798 (to BMZ), and NINDS P30NS047101 (to UCSD microscopy Core). ET was supported by Swiss National Science Foundation Fellowships P2GEP3_155692 and P300PA_164717. TT acknowledges support from a Japan Society for the Promotion of Science (JSPS) for a research abroad fellowship and postdoctoral fellowship (18J00995), and Uehara Memorial Foundation for research abroad fellowship.

Senior Editor

James L Manley, Columbia University, United States

Reviewing Editor

Alan G Hinnebusch, Eunice Kennedy Shriver National Institute of Child Health and Human Development, United States

Publication history

Received: April 13, 2020
Accepted: July 23, 2020
Version of Record published: August 7, 2020 (version 1)

Copyright

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

5,966

Page views
657

Downloads
21

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)

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

Mendeley

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

Tatsuhisa Tsuboi
Matheus P Viana
Fan Xu
Jingwen Yu
Raghav Chanchani
Ximena G Arceo
Evelina Tutucci
Joonhyuk Choi
Yang S Chen
Robert H Singer
Susanne M Rafelski
Brian M Zid

(2020)

Mitochondrial volume fraction and translation duration impact mitochondrial mRNA localization and protein synthesis

eLife 9:e57814.

https://doi.org/10.7554/eLife.57814

Categories and tags

Research organism

S. cerevisiae

Share this article

Cite this article

mRNA association with mitochondria differs between fermentative and respiratory conditions.

Z-projection of live imaging of mRNA and mitochondria in cycloheximide and 1,10-phenanthroline treated cells.

Z-projection of live imaging of mRNA and mitochondria in fermentative condition.

Mitochondrial volume fraction correlates with mRNA localization.

Increased protein synthesis and mRNA localization is regulated by the downstream coding sequence.

Decreased translational elongation localizes mRNA to mitochondria.

Decreased translational elongation localizes mRNA to mitochondria.

Figure 5—source data 1

mRNAs of conserved ATP synthetase components are localized to mitochondria in respiratory and CHX conditions.

mRNA localization to mitochondria enhances its translation.

Mitochondrial volume fraction controls protein synthesis of nuclear-encoded mitochondrial proteins.

Tethering to plasmamembrane reduce protein production in vegetative and fermentative condition.

Author details

Tatsuhisa Tsuboi

Contribution

For correspondence

Competing interests

Matheus P Viana

Present address

Contribution

Competing interests

Fan Xu

Contribution

Competing interests

Jingwen Yu

Contribution

Competing interests

Raghav Chanchani

Contribution

Competing interests

Ximena G Arceo

Contribution

Competing interests

Evelina Tutucci

Present address

Contribution

Competing interests

Joonhyuk Choi

Contribution

Competing interests

Yang S Chen

Contribution

Competing interests

Robert H Singer

Contribution

Competing interests

Susanne M Rafelski

Present address

Contribution

For correspondence

Competing interests

Brian M Zid

Contribution

For correspondence

Competing interests

Downloads (link to download the article as PDF)

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

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

Categories and tags

Research organism

Further reading