Read Me

Read me

Hypoxia associated modulation of the RNA-protein interactome in differentiated kidney tubular epithelial cells

Michael Ignarski, Constantin Rill, Rainer Kaiser, Madlen Kaldirim, René Neuhaus, Reza Esmaillie, Xinping Li, Corinna Klein, Katrin Bohl, Maike Petersen, Christian K. Frese, Martin Hoehne, Ilian Atanassov, Markus M. Rinschen, Katja Höpker, Bernhard Schermer, Thomas Benzing, Christoph Dieterich, Francesca Fabretti and Roman-Ulrich Müller#

# To whom correspondence should be addressed: Roman-Ulrich Müller, Dept. II of Internal Medicine, Kerpener Str. 62, 50937 Köln, roman-ulrich.mueller@uk-koeln.de.


Visualization of the supplemental table to Ignarski et al. 2018 is presented in 3 tabs:

  • RBPome: Volcano plot for the RBPome in mIMCD3 cells under normoxic conditions.
  • Proteome Normoxia: iBAQ ranks of proteins quantified in the mIMCD3 proteome (mean iBAQ over all samples).
  • Proteome Hypoxia vs. Normoxia: Volcano plot of mIMCD3 proteome normoxia vs hypoxia (log2(foldchange) and -log10(p-value) of contrast hypoxia versus normoxia in proteome dataset) and correlation plot of RBPome in normoxic and hypoxic conditions.

On each of the three tabs the user can choose one of the following "modes":

  • Mark one protein (choose one of the three options):
    • Choose the gene name from the drop down menu. OR
    • Type the gene name in the "Show a protein" field. OR
    • Select a protein in the plot (left mouse click).
  • Mark all class I proteins:
    • Class I proteins are defined as RBPs that are significantly enriched (based on iBAQ intensities) in the crosslinked samples over non-crosslinked samples (t-test, FDR<0.1; s0=0.1) in either one of the three t-tests (total dataset, hypoxia, normoxia). Furthermore, all proteins, for which significance could not be calculated due to lacking detection in non-crosslinked samples (identification in ≤ 1 out of 6 replicates) despite clear presence after crosslinking (identification in ≥ 4 out of 6 replicates), were attributed to class I. 510 proteins fulfilled these criteria.
  • Mark all class II proteins:
    • Class II proteins are defined as proteins that were not significantly enriched but that had previously been identified to show RNA-binding capacity based on the recently published compendium, that includes 6 previously published datasets for Mus musculus and 6 for Homo sapiens (Hentze et al., 2018).
  • Mark all novel RBPs:
    • RNA-binding proteins are classified as novel RBPs as the 25 class I RBPs that have been identified in our RIC that have not been reported in previous screens of the mammalian RBPome.
Below each plot the data table is displayed. The data table for preselected proteins can be downloaded with the button "Download data table". For the full data tables for proteome and RBPome refer to the supplementary material.

Please note:

  • The visualized values in the RBPome plot are based on the total dataset t-test. Consequently, proteins reaching significance in only hypoxic or normoxic condition may appear below significance threshold in this plot.
  • More than one dot in plot for one gene name: Sometimes more than one measurement is assigned to one gene name. Usually these are cases with differing protein IDs (probably isoforms etc.) associated with one gene name (see data table).
  • No dot assigned to a gene name in a plot: Some proteins are only measured in the proteome or in the RBPome. These proteins are missing in the respective other plot. The drop down menu contains all proteins.
  • Proteins that are not measured at all in either -CL or +CL are not included in the plot. For these proteins please refer to the supplementary table.

Raw data:

Raw data and MaxQuant (version 1.5.3.8) output were uploaded to the PRIDE repository (http://www.ebi.ac.uk/pride; Project accession: PXD010530)
info1

Volcano plot of the RBPome of mIMCD3 cells

Volcano plot for the RBPome in mIMCD3 cells under normoxic and hypoxic conditions. The log2 fold change (based on iBAQ values, cross-linked vs non-cross-linked) on the x-axis, the significance (-log10 p-value) on the y-axis.

Please note:

  • The visualized values in the RBPome plot are based on the total dataset t-test. Consequently, proteins reaching significance in only hypoxic or normoxic condition may appear below significance threshold in this plot.
  • Proteins with a p-value of zero in this plot were defined as RBPs by the following criterium: identification in ≥ 4 out of 6 crosslinked replicates and identification in ≤ 1 out of 6 non-crosslinked replicates.

Table:

The columns of the table below the plot show:

  • Uniprot Ids and Gene name show the Uniprot Identifier and the gene name of the protein, respectively.
  • Quantified in proteome and quantified in RBPome show a "+" sign if the protein was quantified in the proteome or in the RBPome, respectively.
  • t-test total dataset, t-test Hypoxia and t-test Normoxia show a "+" sign if the protein is significant (FDR<0.1, s0=0.1), "n.s." (not significant) otherwise, in the following contrasts:
    • Total dataset: +CL versus -CL.
    • Hypoxia: +CL versus -CL for hypoxia.
    • Normoxia: +CL versus -CL for normoxia.
  • Class shows the class (I, II, or no entry) of the protein.
  • Novel RBP shows a "+" sign if the protein is categorized as a novel RBP.
info2

iBAQ ranks of proteins quantified in the mIMCD3 proteome.

iBAQ rank plot of the mIMCD3 proteome as a mean over all samples. The data table also lists proteins not quantified in the proteome.

info3

Proteome volcano plot and RBPome correlation

Left plot: Proteome volcano plot

The left plot shows the log2 fold change on the x-axis (based on iBAQ values), and the significance (-log10 p-value) on the y-axis for the proteome for the comparison hypoxia versus normoxia.

Right plot: RBPome correlation plot

Scatter plot showing the correlation of hypoxia log2 fold changes (+CL versus -CL) on the x-axis versus the log2 fold change values of normoxia (+CL versus -CL) on the y-axis.

The linear regression (y = 0.3170 + 0.7069*x) is shown as a black line. The grey lines indicate the 95% prediction interval above and below the linear regression.

Table:

The columns of the table below the plots show:

  • Uniprot Ids and Gene name show the Uniprot Identifier and the gene name of the protein, respectively.
  • t-test Hypoxia vs Normoxia shows a "+" sign if the protein is significant in contrast hypoxia versus normoxia (proteome dataset, FDR<0.1, s0=0.1), "n.s." (not significant) otherwise.
  • Log2FC Hypoxia vs Normoxia shows the log2(foldchange) of contrast hypoxia versus normoxia.