Background

Throughout the tree of life (TOL), ribosomes play a pivotal role in synthesizing all proteins by decoding mRNA [3]. The most precise representations of the TOL to date have been crafted using aligned genomic sequences of ribosomal RNA (rRNA) and ribosomal proteins (r-proteins) [4–7]. This centrality underscores the significance of ribosomal components, many of which feature prominently in the universal gene set of life (UGSL) [8]. The UGSL comprises orthologous genes conserved across the phylogenetic TOL, altered only through speciation while retaining their fundamental functions [9–11]. The interaction of r-proteins and rRNAs throughout the TOL has necessitated a conserved, basic, structural riboproteome. In opisthokonts, the complex has significantly increased in size, both in terms of rRNA expansion segments and in number of associated r-proteins [3]. The r-proteins themselves have become more complex and, in many cases, acquired new functions. Among acquired functionality, ribosomes have gained the ability to selectively decide which transcripts to translate, a concept known as ribosome specialization [12,13]. For example, ribosomal populations enriched in RPL10 translate transcripts subsets [14], and ribosomes lacking RPS26 preferentially translate transcripts from stress-response pathways [15]. To what extent and what proportion of the r-proteome has acquired these types of adaptations remains unknown. Nevertheless, it is becoming increasingly clear that the composition of riboproteomes at a given time reflects structural and functional features of ribosomes [15–17] and grants them the ability to control translational output and shape protein expression.

The field of structural systems biology has progressed greatly in recent years due to the convergence of atomic structures, which provide an average scaffold of macromolecular complexes, and omics studies, which elucidate the relative or absolute abundances of their structural components in vivo [18]. Ribosomes have driven the convergence of these methods due to their ubiquitous and essential nature in the cellular environment. Ribosome structures can currently be highly resolved, reaching near-atomic resolution [19–21]. On the other hand, omics methods lag in providing tailored solutions to fully decipher the general status of riboproteomes. Although methods such as transcriptomics or translatomics can monitor gene expression in a significant percentage of the expressed genome, a gap remains between transcriptional regulation and the abundance of active protein pools [22]. Recent efforts include advanced mass spectrometry methods to elucidate different features of r-proteins [23], such as absolute quantification using synthetic peptide standards [14]. These examples make it clear that multiplexing methods will help to overcome the limitation of studying riboproteomes at the ribosome population level.

We focused on three major obstacles that currently prevent the thorough investigation of r-proteins. First, multiplexing requires the purification of riboproteomes in a pseudo-native state that conserves protein features to elucidate different aspects of their biology. Second, the analysis of r-proteins following protease digestion requires careful consideration of the amino acid composition, which includes many basic peptide stretches required for interacting with rRNA. Finally, many translational studies often fail to bridge the gap between translatome and proteome. Currently, the best method to monitor the translational output from ribosomes is Ribo-Seq [24], which has been optimized in plant systems [25] but fails to provide measurements of de novo synthesized proteins. Moreover, typical translational studies deal with organisms that transition between biological steady states. These transitions complicate the use of technologies such as in vivo stable-isotope labeling and mass spectrometry to monitor protein synthesis, as quantification of tracer incorporation assumes steady-state conditions.

Here, we set out to provide a standardized methodology to overcome these three limitations while addressing a previously unsolved biological question in plant translation [1]. To deal with the first limitation, we developed a strategy to purify the r-proteome as close as possible to its native state by using a chaotropic agent to dissociate r-proteins from rRNA. Our purification strategy enhances both top-down and bottom-up proteomics methods, as previously described by others [23], by capturing r-proteins with paramagnetic SP3 beads [26–28]. To bypass the second limitation, we carefully consider our choice of an optimal protease to digest ribosomal proteins across the TOL. We validate that our protease yields larger peptides and thus allows for a more comprehensive profiling of the r-proteome from bacteria to higher metazoans. To address the third limitation, we provide an R package called ProtSynthesis (https://github.com/MSeidelFed/ProtSynthesis), in which we developed physiological and data-based assumptions required to quantify tracer incorporation and fractional protein synthesis rates in organisms shifting between biological states, using plant cold acclimation as an example [1,29,30]. Our method uses organismal physiology in the treated vs. non-treated condition to make fractional synthesis rates comparable. First, we quantify the enrichment percentage in soluble amino acid pools and use them to constrain the maximum amount of nitrogen atoms that could have incorporated the label in specific peptide sequences, thus correcting for biases in ¹⁵N incorporation into soluble amino acid pools across conditions. Secondly, treatments often induce differences in growth, and we calculate relative growth rates to normalize fractional protein synthesis. Finally, because the most direct measurement of translation is protein content, we correct our relative growth rates to reflect the percentage of protein accumulated per unit of time. In summary, our calculations comprehensively incorporate organism physiology to enhance the comparison between fractional protein synthesis rates across different biological steady states, such as cold-acclimated plants vs. plants reared at optimal temperatures. Beyond cold acclimation, there are plant examples in which biological steady state shifts occur concomitantly to surmised translational reprogramming, which could profit from our methodology. For instance, 3',5'-cAMP supplementation induces spatially constrained rearrangements of the r-proteome in Arabidopsis [31]. Similarly, flg22 induces dissociation of the P-Stalk via the activation of Mitogen-activated protein kinase 6 pathway, influencing plant immunity to bacteria [16]. These examples set an ideal stage for inquiring about changes in the synthesis and assembly of the r-proteome and how it affects protein synthesis, i.e., what is the functional translational implication of remodeling the r-proteome. To summarize, we provide a comprehensive methodology to couple the most accurate methodology to study translational efficiencies (Ribo-Seq) to r-protein and proteome-wide fractional protein synthesis rates.

Summary

Our method contains five steps that are depicted in the graphical abstract. Step 1a entails labeling the biological specimens in a manner tailored to each organism. The required methodology, including materials and labeling considerations for plants, is discussed in detail elsewhere [2]. Step 1b entails a detailed phenotypic analysis of the organism of interest upon applying a specific treatment that produces a shift from the previous biological steady state. For phenotyping and growth analyses of plants, we use and recommend the classical definition of relative growth rates that has been discussed extensively in the literature over the last century [32–34] based on the first postulation of an efficiency index by Blackman, 1919 [35], which is equivalent to the relative growth rate. Step 2a accounts for differential labeling efficiency in soluble amino acid pools at the onset of the investigated physiological transition. The methodology for obtaining the primary metabolome and measuring metabolite abundances by gas chromatography–mass spectrometry has been described elsewhere [2]. The contents of the manuscript detail the procedure to purify and profile a complexOme or ribosomal proteome (Step 2b) and elaborate on the data analyses and further considerations needed to calculate protein fractional synthesis rates from protein components of isolated complexes (Step 3).