(A) The six different r-protein assembly trees (initiated by primary binding r-proteins) of the E. coli 30S subunit are ordered according to their physical location on the 16S rRNA (in 5′ to 3′ direction) and attributed to 16S rRNA domain organisation (5′, central, and 3′ domain). The r-proteins are classified by their binding hierarchy. Primary binding proteins (1°) are capable of initiating pioneering interactions with rRNA independent of other proteins. The secondary binders (2°) require one or more primary binding proteins for their association with rRNA, while tertiary binding (3°) proteins require both primary and secondary binders for their incorporation into ribosomal subunits. If existing, homologous r-proteins in S. cerevisiae (rpS nomenclature) are shown next to their prokaryotic counterparts. (B) A schematic presentation of the tertiary structure of the 16S rRNA is depicted. Each of the three major secondary structure domains of the 16S rRNA forms distinct morphological features of the 30S subunit. The 16S rRNA 5′ domain forms the shoulder and foot (red), the central domain forms the platform (green) and the 3′ major domain forms the head (blue). The assembly map of (A) is superimposed in this schematic structure visualisation paying attention to the localisation of the respective r-protein. The colour of the circle gives information about the assembly hierarchy of the respective r-protein (see also A). S11/rpS14 is classified in a species-dependent manner as a tertiary binder (E. coli) or a primary binder (Aquifex aeolicus) [60]. Only r-proteins with sequence homologous in S. cerevisiae (rpS nomenclature) are shown. The figure is reproduced and adapted from [4] (adaptation from the original assembly map of Nomura and colleagues [1]).

The indicated yeast strains carrying galactose inducible alleles of the indicated SSU r-protein genes in combination with TAP-tag fusion alleles of UTP-A component Utp4p (A), or UTP-B component Pwp2p (B), were either cultivated in medium containing galactose (Gal) as carbon source or were transferred to glucose containing medium (Glu) and cultivated for additional four hours to turn off the expression of the respective r-proteins. TAP-tagged bait proteins were affinity purified via their Protein A moiety using IgG sepharose beads. The amount of purified bait protein was monitored by Western blotting (lower panels) and co-purified pre-rRNA species were analysed by Northern blotting (upper panels) using oligo 1819, which hybridizes in ribosomal precursor rRNAs between 18S and 5.8S rRNA sequences and detects 35S, 32S, 23S, and 20S pre-rRNAs (see Fig. S1). Equal signal intensities of input (In) and beads (IP) fractions in Northern blots correspond to 1% co-precipitation of the respective rRNA. Efficiencies of 35S pre-rRNA purification normalized to the values obtained for cells grown in permissive conditions are indicated in the lower panels. For the Western blot analyses equal signal intensities of input (In) and beads (IP) correspond to 20% precipitation of the TAP-tagged bait protein. The strains are ordered in regard to the binding of the respective r-proteins to the three major secondary structure domains of the 18S rRNA. Prokaryotic homologues of rpS11, rpS9, rpS22, rpS13, and rpS5 are primary rRNA in vitro binders. Prokaryotic homologues of rpS15 and rpS14 are secondary/tertiary in vitro binders of the assembly trees initiated by binding of the homologues of rpS13 and rpS5, respectively (see Fig. 1).

The yeast strain TY1907 (wildtype) expressing chromosome encoded TAP tagged Utp4p, and conditional mutant yeast strains expressing chromosome encoded TAP tagged Utp4p and carrying in addition galactose inducible alleles of RPS5 (TY1524), RPS13 (TY1893), or RPS14 (TY2104) were cultivated in medium containing galactose as carbon source and were subsequently transferred to glucose containing medium and cultivated for additional four hours. Utp4p-TAP was affinity purified from corresponding cellular extracts using IgG coupled magnetic beads. Affinity purified proteins were digested by trypsin and the resulting peptides from each sample were labelled with specific iTRAQ reagents. Labelled peptides of wildtype samples were combined with labelled peptides of samples derived from the conditional mutants of either RPS5, RPS13, or RPS14 and were then further analyzed by LC-MS/MS as described in material and methods. Datasets of in total eight (mutant:wildtype) comparisons were generated. In experiments 1–4 Utp4p-TAP fractions purified from the wildtype strain (TY1907) were compared with Utp4p-TAP fractions purified from the conditional RPS5 mutant (TY1524). In experiments 5–7 Utp4p-TAP fractions purified from the wildtype strain (TY1907) were compared with the ones purified from the conditional RPS13 mutant strain (TY1893). In experiment 8 Utp4p-TAP fractions from the wildtype strain were compared with the one purified from the conditional RPS14 mutant strain (TY2104). Experiments 3 and 6 are duplicates of the LC-MS/MS analysis of experiments 2 and 5, respectively. iTRAQ ratios of SSU processome components identified in 5 or more of the 8 experiments were combined to one dataset and statistical clustering algorithms were applied as described in material and methods. (A) shows a comparison of the similarity of the eight individual experimental datasets in regard to each other and (B) shows a clustering analysis of the identified SSU processome components in regard to their iTRAQ ratios in the eight experiments. In (B) boxes in red colours represent relative enrichment and boxes in green colours relative deprivation of a protein in Utp4p-TAP fractions purified from mutant versus wildtype cells. Boxes in gray colour indicate that no peptide of the respective protein could be identified in the corresponding experiment. Standard names of the identified components are indicated in (B) on the left. On the right the major protein groups described in the text are designated and it is indicated if a component belongs to the UTP-A, UTP-B, or UTP-C SSU processome sub-module. We note that despite the overall highly similar composition of Utp4-TAP purifications from cells in vivo depleted of rps13 (experiments 5–7) or rpS14 (experiment 8) the dataset gave first indications for specific differences between these pre-ribosomal populations, as in the content of Krr1p.

The yeast strains carrying galactose inducible alleles of the indicated SSU r-protein genes in combination with TAP-tag fusion alleles of Noc4p were either cultivated in medium containing galactose (Gal) as carbon source or were transferred to glucose containing medium (Glu) and cultivated for additional four hours. Noc4p-TAP was affinity purified via its Protein A moiety using IgG sepharose beads. The amount of purified Noc4p-TAP was monitored by Western blotting (lower panels) and co-purified pre-rRNA species were analysed by Northern blotting (upper panels) using oligo 1819, which hybridizes in ribosomal precursor rRNAs between 18S and 5.8S rRNA sequences and detects 35S, 32S, 23S, and 20S pre-rRNAs (see Fig. S1). Equal signal intensities of input (In) and beads (IP) fractions in Northern blots correspond to 1% co-precipitation of the respective rRNA. Efficiencies of 35S pre-rRNA purification normalized to the values obtained for cells grown in permissive conditions are indicated in the lower panel. For the Western blot analyses equal signal intensities of input (In) and beads (IP) correspond to 20% precipitation of the TAP-tagged bait protein. The strains are ordered in regard to the binding of the respective r-proteins to the three main secondary structure domains of the 18S rRNA. Prokaryotic homologues of rpS11, rpS9, rpS22, rpS13, and rpS5 are primary rRNA in vitro binders. Homologues of rpS15 and rpS14 are secondary/tertiary in vitro binders in the assembly trees initiated by binding of the homologues of rpS13 and rpS5, respectively (see Fig. 1).

The yeast strain TY96 (wildtype) expressing chromosome encoded TAP tagged Noc4p and yeast conditional mutant strains TY1241, TY1897, and TY1902, expressing chromosomal encoded TAP tagged Noc4p and carrying in addition galactose inducible conditional alleles of RPS5 (TY1241), RPS13 (TY1897), or RPS22 (TY1902) were cultivated in medium containing galactose as carbon source and were subsequently transferred to glucose containing medium and cultivated for additional four hours. Noc4p-TAP was affinity purified from corresponding cellular extracts using IgG coupled magnetic bead matrix. Affinity purified proteins were digested using trypsin and the resulting peptides from each sample were labelled with different iTRAQ reagents. Labelled peptides of wildtype samples were combined with labelled peptides of samples derived either from the conditional mutant of RPS5 (A), RPS13 (B), or RPS22 (C) and were then further analyzed as described in material and methods. Average iTRAQ ratios of each SSU processome component identified by more than one peptide are indicated in (A)–(C). Numbers in brackets behind SSU processome component names indicate the number of peptides (confidence interval >95%) by which the respective protein was identified. (D) shows a heatmap representation of the three datasets. The factors are ordered according to a clustering analysis (see material and methods). Boxes in red colours represent relative enrichment and boxes in green colours relative deprivation of a protein in Noc4p-TAP fractions purified from mutant versus wildtype cells.

The temperature sensitive noc4–8 yeast mutant strain (TY40) was transformed with vectors supporting the constitutive expression of Flag tagged SSU r-proteins (see Fig. S4). Overnight cultures of transformants were grown for one generation time in full medium at 24°C to an OD of 0.4 and then cultivated for three hours in full medium at either permissive (24°C) or restrictive (37°C) temperature. The respective Flag-tagged r-protein was affinity purified from cellular extracts using anti-Flag M2 beads and co-purifying (pre-) rRNA species were analysed by Northern blotting using oligo 1819, which hybridizes in ribosomal precursor rRNAs between 18S and 5.8S rRNA sequences and detects 35S, 32S, 23S, and 20S pre-rRNAs (see Fig. S1). Oligo 205, which hybridizes within the 18S region, was used to detect 18S rRNA. Equal signal intensities of input (In) and affinity purified (IP) fractions correspond to 3% co-purification of the respective rRNA. The numbers in the lower panels indicate the efficiencies of 20S pre-rRNA purification divided by the efficiencies of 18S rRNA purification to normalize for possible over-all variations in the individual immuno-purification experiments. However, we note that the changes in 18S rRNA co-purification efficiencies between experiments performed with one transformant grown in permissive versus non-permissive conditions were in general below twenty percent. The numbers shown are the average of the results of two to four independent experiments and the values obtained with cells grown in permissive conditions were set to one.