A comparative study of ribosomal proteins: linkage between amino acid distribution and ribosomal assembly

Abstract

Background: Assembly of the ribosome from its protein and RNA constituents must occur quickly and efficiently in order to synthesize the proteins necessary for all cellular activity. Since the early 1960's, certain characteristics of possible assembly pathways have been elucidated, yet the mechanisms that govern the precise recognition events remain unclear.We utilize a comparative analysis to investigate the amino acid composition of ribosomal proteins (r-proteins) with respect to their role in the assembly process. We compared small subunit (30S) r-protein sequences to those of other housekeeping proteins from 560 bacterial species and searched for correlations between r-protein amino acid content and factors such as assembly binding order, environmental growth temperature, protein size, and contact with ribosomal RNA (rRNA) in the 30S complex.

Results: We find r-proteins have a significantly high percent of positive residues, which are highly represented at rRNA contact sites. An inverse correlation between the percent of positive residues and r-protein size was identified and is mainly due to the content of Lysine residues, rather than Arginine. Nearly all r-proteins carry a net positive charge, but no statistical correlation between the net charge and the binding order was detected. Thermophilic (high-temperature) r-proteins contain increased Arginine, Isoleucine, and Tyrosine, and decreased Serine and Threonine compared to mesophilic (lower-temperature), reflecting a known distinction between thermophiles and mesophiles, possibly to account for protein thermostability. However, this difference in amino acid content does not extend to rRNA contact sites, as the proportions of thermophilic and mesophilic contact residues are not significantly different.

Conclusions: Given the significantly higher level of positively charged residues in r-proteins and at contact sites, we conclude that ribosome assembly relies heavily on an electrostatic component of interaction. However, the binding order of r-proteins in assembly does not appear to depend on these electrostatics interactions. Additionally, because thermophiles and mesophiles exhibit significantly different amino acid compositions in their sequences but not in the identities of contact sites, we conclude that this electrostatic component of interaction is insensitive to temperature and is not the determining factor differentiating the temperature sensitivity of ribosome assembly.

Figure 1

Student’s T-test shows significant differences between ribosomal and non-ribosomal proteins. (A) Average amino acids compositions found in ribosomal proteins (purple) and non-ribosomal proteins (light blue) samples. Asterisks indicate a statistically significant difference between the two averages; error bars are ±σ. (B): T-values from Student’s t-tests for the amino acids compositions with significant difference between the two groups. The magnitude of the bar represents the relative difference between the two means and the direction of the bar (up or down) indicates which protein sample contains the larger proportion of that residue. A positive T-value indicates a higher proportion of that residue found in the non-ribosomal sample, whereas a negative T-value corresponds to a higher proportion of that residue found in ribosomal proteins.

Figure 2

R-protein amino acid compositions exhibit typical thermostability differences. Thermophilic r-proteins (red) contain higher percentages of Arginine (R), Isoleucine (I), and Tyrosine (Y), and lower percentages of Serine (S) and Threonine (T) than mesophilic r-proteins (blue). These differences are generally consistent with typical differences among thermophilic and mesophilic proteins and are estimated to function in the thermostability of the protein. In the box-and-whisker representation, the lower and upper circles represent the 5th and 95th percentiles, respectively, and the lower and upper whiskers the 10th and 90th. The colored regions mark the middle 50% of the samples (25th to 75th percentile), with a solid line representing the median and a dotted line the mean. Asterisks mark the amino acids that show a statistically significant difference between mesophilic and thermophilic species.

Figure 3

Percentage of positively charged residues correlates with protein size but net charges do not. All thermophilic r-proteins except S11 contain a higher percentage of positively charged residues than their mesophilic homologs (A), and, for some proteins, including all six primary proteins, this difference is statistically significant. R-proteins generally have a net positive charge (B), and thermophiles typically have a higher average charge than mesophiles. For three proteins, this difference is significant. The box-and-whisker plots are represented as in Figure 2.

Figure 4

Proportion of Lysine residues correlates with average protein length but Arginine does not. Lysine (A) shows a highly significant negative correlation with protein length (Spearman’s rank correlation: ρ = −0.802, p=2.60x10^-5), whereas Arginine (B) shows a weaker correlation with no statistical significance (ρ = −0.484, p=0.032).

Figure 5

Generally, contact residue identities are not statistically different between mesophiles and thermophiles. (A) R-proteins (purple) show reasonable distributions of amino acids at contact sites: positively charged and polar residues are likely to interact with the negatively charged rRNA, so high CEF are expected. A CEF > 1 indicates a high prevalence for that amino acid to be located at a contact site; a CEF < 1 indicates a deficiency; a CEF ~ 1 indicates no preference for that amino acid to be located at contact versus non-contact sites. Asterisks indicate the residues whose CEF deviate significantly from 1 (Student’s t-test, α=0.01). The box-and-whisker plots are represented as in Figure 2. (B) CEF for the amino acids at the estimated rRNA contact sites for mesophilic (blue) and thermophilic (red) r-proteins. Only Glutamic acid, E, shows a CEF mean that is statistically different between mesophiles and thermophiles. This is an intriguing finding, considering the significant differences in overall amino acid composition, as shown in Figure 2.