Figure S1
Click on the picture to view a high resolution version
Correlation SAM analysis - Mitochondrial localization rate
Histograms of the three classes of transcripts determined after SAM analysis: Localized (enriched in the mitochondrial fraction), Undetermined (not enriched or not reduced in the mitochondrial fraction); Not localized (specifically reduced in the mitochondrial fraction). The top histogram represents the distribution of transcripts with a calculated percentage of mitochondrial localization value (MLR) superior (blue) or inferior (red) to 10%. The bottom histogram represents the same distribution, but with a calculated percentage of mitochondrial localization superior (blue) or inferior (red) to 8%. The number in the bars represents the quantity of transcript that belongs to the different classes with the proportion of these classes on the right side of the bars. It is important to note that almost all the transcripts (96,3%) with a percentage of localization superior to 8% belongs to the "localized" class, showing a high correlation between SAM analysis and the calculated percentage of mitochondrial localization
Figure S2
Click on the picture to view a high resolution version
FISH images of class I, I, III nuclear-encoded mRNAs coding for mitochondrial proteins
Five fluorescent DNA probes specific for mitochondrial ribosomal RNA delimit the mitochondrial compartment (red). mRNAs were labeled with specific fluorescent probes (green). 71 to 190 cells were examined and quantification is represented on a histogram. The quantification using Corsen software (see Materials and Methods, Jourdren et al., in preparation) allows the calculation of the percentage of cells in which a specific mRNA co-localized with mitochondria. The represented confident interval is calculated assuming a binomial distribution. Each classes of transcripts are represented, ATP16 mRNAs (class III) are not localized to mitochondria whereas ATP4 (class II) or BCS1 and YAH1 (class I) mRNAs co-localized with mitochondria.
Figure S3
Click on the picture to view a high resolution version
Is mRNA accumulation induced in Delta-puf3 strain correlated with mRNA delocalization?
To determine a putative correlation between accumulation of class I mRNA and mitochondrial delocalization due to the absence of Puf3p, variation of steady state level of 11 class I mRNAs is plotted against the variation of the MLR value of these transcript in Delat-puf3 versus wild type strain. Pearson correlation value was calculated and shows a poorly significant correlation between accumulation of mRNA and delocalization of mitochondria with a confidence interval of 5%. In addition, SAM50 mRNA was classified as a class I mRNA after it was found as a Puf3p target in studies of Gerber et al (2004, PloS Biol 2 E79), however, SAM50 mRNA has no Puf3p motif I its 3'UTR and, in our study, its MLR value is not affected by PUF3 deletion. Thus SAM50 may not be a class I mRNA and when removed from the above list, the corresponding Pearson value shows no correlation between mRNA accumulation and mitochondrial delocalization. In conclusion, the half-life of more mRNAs should be analyzed to establish the pivotal role of Puf3p in localization-translation-degradation of all these class I mRNAs.
Figure S4
Click on the picture to view a high resolution version
High salt concentration does not lead to dissociation of mRNAs from the mitochondria.
We analyzed by quantitative RT-PCR the amount of 9 mRNAs co-purified with mitochondria in different conditions: 0, 1M KCl, 1M KCl + puromycin. As previously observed [10], the salt effect is not sufficient to free the mRNAs from mitochondria; only the double action of KCl + puromycin is efficient on the majority of mRNAs (see Results and Table S1). Each measure was made in triplicate and a minimum of two independent quantitative PCR has been performed allowing the calculation of a standard deviation presented on the histogram.
Figure S5
Click on the picture to view the video version
3D views of FISH analyses
The image J plugin "Volume Viewer" has been used to reconstruct a 3 dimensional view of mRNA localization (green) with mitochondria (red) in the cells pictured in the Figure S2. The nucleus is stained in blue.
Colocalization of ATP4 a class II mRNA with mitochondria in WT cells.
Figure S6
Click on the picture to view the video version
3D views of FISH analyses
The image J plugin "Volume Viewer" has been used to reconstruct a 3 dimensional view of mRNA localization (green) with mitochondria (red) in the cells pictured in the Figure S2. The nucleus is stained in blue.
ATP16 a class III mRNA is not localized to the vicinity of mitochondria.
Figure S7
Click on the picture to view the video version
3D views of FISH analyses
TThe image J plugin "Volume Viewer" has been used to reconstruct a 3 dimensional view of mRNA localization (green) with mitochondria (red) in the cells pictured in the Figure S2. The nucleus is stained in blue.
YAH1 a class I mRNA colocalizes with mitochondria in WT cells.
Tables
| Table | description |
|---|---|
| Table S1 | MLR properties of the mRNAs coding for mitochondrial products. The genes are grouped in functional modules following the work by Steinmetz group (Perocchi F, Jensen LJ, Gagneur J, Ahting U, von Mering C, et al. (2006). PLoS Genet 2: e170). Four MLR classes of RNA are distinguished by different colours. Class I and class II are mRNAs whose statistical analyses (SAM q values) indicate an asymmetrical localization. This mitochondrial localization is dependent or independent of the mRNA binding protein Puf3p, respectively. Class III corresponds to mRNAs translated on cytoplasmically-free ribosomes and class IV are mRNAs encoded in the mitochondrial genome. The MLR values (% of Mitochondrial Localization of mRNA) are represented for the wild type cell in standard growth conditions or with either cycloheximide or puromycin. The mutant strain Delta-puf3 has also been analyzed. The decision whether a gene has prokaryotic homologue is from SGD database. Puf3p binding site is from the data of (Foat BC, Houshmandi SS, Olivas WM, Bussemaker HJ (2005) Proc Natl Acad Sci USA 102: 17675-17680) and (Gerber AP, Herschlag D, Brown PO (2004) PLoS Biol 2: E79). Puf3p target signifies that the corresponding mRNA has been retained in affinity chromatography with Puf3p (Gerber AP, Herschlag D, Brown PO (2004) PLoS Biol 2: E79). |
| Table S2 | RT-PCR analyses of mRNA localization. RT-PCR analyses were conducted as indicated in Garcia M, Darzacq X, Devaux F, Singer R, Jacq C (2007); Biology MiM, editor. Totowa, New Jersey: Humana Press. 505-528 p. |
| Table S3 | The COX assembly factors are class I MLR-mRNAs. The different classes of genes were described previously (Fontanesi F, Soto IC, Horn D, Barrientos A (2006) Am. J. Physiol. Cell 291:C1129-1147). All the assembly factors are translated to the vicinity of mitochondria and 17 out 18 are likely to interact with Puf3p. The 8 nuclearly-encoded mRNAs coding for structural proteins are translated on free cytoplasmic polysomes. |
| Table S4 | List of the genes coding for cytosolic non-mitochondrial proteins (according to YPD) which were used as a negative control set. |
Microarray data
Microarray data are available from the GEO repository under the series accession number: GSE9393.
Document S1
QUANTITATIVE PCR ANALYSIS OF ASYMMETRIC mRNA LOCALIZATION. A complete version of this protocol has been published in Garcia M et al. 2007. Download the Doc. S1 PDF file.