The Skaggs Institute
for Chemical Biology

Scientific Report 2007

A Postproteomic Approach for Studies of Ribosome Assembly

J.R. Williamson, F. Agnelli, A. Beck., C. Beuck, A. Bunner, A. Carmel, S. Chen, S. Edgcomb, D. Kerkow, E. Kompfner, S. Kwan, E. Menicelli, W. Ridgeway, G. Ring, H. Schultheisz, L.G. Scott, E. Sperling, M. Sykes, B. Szymczyna, J. Wu

The bacterial ribosome is composed of the 50S subunit, responsible for forming peptide bonds, and the 30S subunit, responsible for decoding messenger RNA. Biosynthesis of the 55 small proteins and 3 large ribosomal RNA molecules is carefully choreographed and regulated so that the correct proportions of these many components are produced. The ribosome is responsible for protein synthesis in all cells and therefore is essential to ensure the accurate and efficient assembly of these key protein synthesis factories.

The 30S ribosome is composed of 20 proteins and the 1500-nucleotide 16S ribosomal RNA; this particle serves as the model for macromolecular assemblies. The 30S ribosome can be reconstituted from its purified components in vitro, making it amenable to biochemical and biophysical investigations to understand the mechanism of assembly. Earlier research in other laboratories led to an assembly map that provided the thermodynamic binding order of the 20 proteins to the 16S ribosomal RNA. We have been investigating the kinetics of 30S assembly by using newly developed methods to determine the order of events and the nature of the rate-limiting steps for ribosome assembly in vitro.

We have developed an isotope pulse-chase assay that can be used to monitor the binding kinetics of all 20 ribosomal proteins simultaneously (Fig. 1). The assembly is initiated with ¹⁵N-labeled ribosomal proteins, and after a certain assembly time, the assembly is chased by addition of unlabeled (¹⁴N) proteins. The assembly is allowed to proceed to completion, and the ratio of ¹⁵N to ¹⁴N proteins is determined as a function of the assembly time by using mass spectrometry. With this powerful method, we showed that assembly proceeds from the 5′ end to the 3′ end of the 16S ribosomal RNA, findings consistent with the direction of biosynthesis in vivo. The slow step is the assembly of the 3′ domain, which is the head of the 30S ribosome that undergoes ratcheting motion during the translocation step of protein synthesis.

Fig. 1. Assembly reaction of the 30S ribosomal subunit. A total of 20 proteins (represented as circles) bind to the 16S ribosomal RNA (shown as a line). In the absence of the proteins, the RNA is only partly folded (left), but it becomes highly ordered as it folds during assembly. The complex reaction shown here occurs in a stepwise manner to ensure proper assembly, but the details of this process remain to be elucidated.

In the past year, we markedly extended the accuracy and precision of this method. Our previous analysis relied on matrix-assisted laser desorption mass spectrometry of the intact ribosomal proteins. Although this analytical method provided insights into ribosome assembly, it had poor signal-to-noise ratios and inefficient ionization of the larger proteins. To remedy this problem, we adapted methods from proteomic mass spectrometry to fit our quantitative needs. Using the same isotope pulse-chase assay, before mass spectrometry analysis, we digest the ribosomal proteins with trypsin to generate many small peptides.

The set of ribosomal peptides is analyzed by using liquid chromatography and electrospray ionization time-of-flight mass spectrometry. The peptides are first separated by using reverse-phase high-performance liquid chromatography, and the effluent from the chromatograph is inserted directly into the mass spectrometer so that the mass spectrum of the eluting peptides is recorded as a function of time. A data set from this analysis is shown in Figure 2 as a 2-dimensional contour plot.

Each spot in Figure 2 corresponds to a particular ribosomal peptide, and clear pairs of peaks correspond to the unlabeled (¹⁴N) and labeled (¹⁵N) species. Before this complex data can be used, the peptides must be identified, and they must be accurately integrated to determine the isotope ratios for each peptide. We have developed novel assignment algorithms that enable us to determine the identity of the peptides from the accurate mass, the charge state of the peptide ion, and the number of nitrogen atoms. For quantification, we have developed novel Fourier transform convolution methods in which the high-resolution isotope distribution for each ¹⁴N-¹⁵N peak pair can be quantitatively analyzed. Compared with our previous analysis based on matrix-assisted laser desorption mass spectrometry, this approach has yielded a 4- to 5-fold increase in the precision of the kinetic measurements.

Fig. 2. A 2-dimensional representation of a liquid chromatography plus mass spectrum analysis of unlabeled (14N) and labeled (15N) ribosomal peptides. A, In a contour plot of the data, the black spots correspond to peptides with a particular mass eluting at a particular retention time. The peaks occur in pairs that correspond to the 14N and 15N peptides. B and C, In an expansion of 2 representative peaks, the high-resolution cross section of the peaks reveals the fine structure of the isotope distribution that occurs from natural-abundance isotopes in the peptides. Part of the theoretical trypsin digestion list is shown for each peak, indicating the identity of the peptides according to highly accurate mass measurements. The peptides can be identified because of the known mixture of 30S ribosomal proteins in the sample, which reduces the complexity of the proteomic analysis.

Using this new postproteomic mass spectrometry method, we are further exploring the kinetics of ribosome assembly in vitro. We have purified and expressed all of the individual ribosomal proteins, and we can now perform kinetic isotope pulse-chase experiments with predefined mixtures of subsets of the ribosomal proteins. Use of predefined mixtures enables us to determine the kinetic cooperativity between the ribosomal proteins. We are particularly interested in determining the nature of the slowest steps in the assembly of the 3′ domain.

With this new analytic method, we can also attack the problem of ribosome assembly in vivo. Using the new method directly with ribosomes isolated from cells, we can track the assembly of ribosomes by using pulses of isotopes added directly to living cells. This exciting new direction will enable us to compare our considerable body of work on assembly in vitro with characteristics of the cellular ribosome biogenesis pathway. We are extending this analysis from assembly of bacterial ribosomes to the other major classes of ribosomes, and we are actively pursuing assembly of ribosomes from yeast, plant chloroplasts, and human cells.

Publications

Karnaukhov, A.V., Karnaukhova, E.V., Williamson, J.R. Numerical matrices method for nonlinear system identification and description of dynamics of biochemical reaction networks. Biophys. J. 92:3459, 2007.

Vallurupalli, P., Scott, L., Williamson, J.R., Kay, L.E. Strong coupling effects during X-pulse CPMG experiments recorded on heteronuclear ABX spin systems: artifacts and a simple solution. J. Biomol. NMR 38:41, 2007.