...and RNA makes Protein. (part 1)
Note. This is a longread (2,100 words) with a calculated 11 min read time, not counting the time it takes you to look closely at the figures and read the captions.
by Christoph
You read the title correctly as the second part of the 'central dogma of molecular biology', casually expressed as "DNA makes RNA, and RNA makes protein". In its original formulation, Francis Crick focused on the flow of information from DNA to protein, the "detailed transfer of sequence information from one unit to the next" (ref.). He did not go into detail about the biochemical reactions of and between the 'units,' that is, the molecules involved.
Figure 1. Crick's first outline of the central dogma, from an unpublished note made in 1956. Source
But that is exactly what I intend to do here: take the title "… and RNA makes protein" literally and describe what researchers have figured out about where and how RNA catalyzes the addition of another amino acid to a growing peptide chain. In more scientific terms: where and how ribosomal RNA catalyzes peptide-bond formation, also known as transpeptidase or peptidyltransferase in enzymology (ENZYME entry EC 2.3.2.12). As a chemical equation, it looks like: peptidyl-[tRNA] + charged tRNA → peptidyl-aminoacyl-[tRNA] + uncharged tRNA [ref.]. Of course, the equation is not at all intuitive and does little to aid understanding – nor is it meant to.
Not a digression, just a brief look back at the late 1970s. In his recent essay The controls that got out of control, André Schneider gives two examples of how failed control experiments – negative controls turned into experiments – paved the way to transformative discoveries. In 1978, Thomas Cech discovered that an intron in the rRNA of Tetrahymena can self-splice without the involvement of protein(s), and Sydney Altman found that the RNA component of RNase P, RnpB, can process the 5'-leader sequence of precursor tRNAs on its own, without the protein component RnpA. Both findings contradicted the prevailing view at the time that enzymatic activity was exclusive to proteins, think of "one gene – one enzyme" (Beadle & Tatum (1941)). The widespread acceptance of catalytically active RNA by the mid-1980s was certainly aided by its discovery in two different experiments at the same time. Schneider (2026) concludes this section of his essay with: "...it was later shown that rRNA catalyzes ribosomal peptide-bond formation, arguably one of the most important biochemical reaction[s]."
That "it was later shown..." struck a chord with me, since in the mid-1980s, as a PhD student at the MPI MolGen in Berlin, Germany, I witnessed the high-tension debates about the nature of ribosomal peptide-bond formation, the elusive peptidyltransferase. We in the Abteilung Wittmann – arguably one of the epicenters of ribosome research back then – took it for granted that the three ribosomal RNAs (16S, 23S, 5S) in the 70S ribosome of E. coli provide the structural framework for its ~55 ribosomal proteins, which, in the right three-dimensional configuration and together with the translation factors EF-Tu and EF‑G, carry out the enzymatic reactions that occur during translation of mRNA via charged tRNAs. This assumption was supported by the work of Luigi Gorini, who showed that the proofreading capacity of ribosomes, their translational fidelity, involves interactions between the three small subunit proteins S12, S4, and S5 (Ref.).
And then Harry Noller burst into department seminars – I vividly recall two such encounters – and put forward his view that the 23S rRNA of the large subunit catalyzes peptide-bond formation, rather than any of the large subunit (L) proteins. All this was based on preliminary results from continuously refined experiments of his lab, later published under the title Unusual Resistance of Peptidyl Transferase to Protein Extraction Procedures (1992). We were skeptical of Noller's results, since they were obtained using a highly artificial in vitro test system and under non-physiological conditions – but having "digested" the findings of Cech and Altmann on catalytic RNA left us, most of us, with the impression that Noller was on the right track after all.
That's all history now, and I could simply point out which nucleotides of the 23S rRNA make up the ribosomal peptidyltransferase (PT). You can see that here. However, I would like to take you on a visual journey inside the ribosome to show you as "tangibly" as possible where ribosomal RNA catalyzes the formation of peptide bonds. First a diagram, a "tourist map" so to say (Figure 2).
Figure 2. Schematic of Peptide Bond Formation on the Ribosome. See text for details. The 50S subunit, where the peptidyltransferase (PT) center is located, is shown in light gray and the 30S subunit in dark gray. A, P, and E sites of the ribosome are indicated. Source
The "tourist map" explains to you peptide-bond formation in the caption: "The α‑amino group of aminoacyl-tRNA in the A site (red) attacks the carbonyl carbon of the pept-tRNA in the P site (blue) to produce a new, one amino acid longer pept-tRNA in the A site and a deacylated tRNA in the P site." This happens within the 50S large subunit, and the 3'-CCA tails of both the A site-bound aminoacyl-tRNA and the P site-bound pept(idyl)-tRNA have to be in close contact for the reaction to occur between their attached amino acids, resp. peptide. The anticodon loops of both tRNAs interacting with their adjacent codons on the mRNA bound to the 30S subunit are irrelevant for the actual peptidyltransferase reaction, as is the E site-bound tRNA.
Note: If the abstract diagram in Figure 2 is enough to help you understand the where of the peptidyltransferase reaction, you can skip the detailed "terrain exploration" that follows and jump straight to part 2 for the how.
To help you gain a more precise spatial understanding of exactly where in the ribosome the peptidyltransferase reaction takes place, prepare for a move from the flat, 2D diagram in Figure 2 to 3D structures – necessarily shown below as 2D images taken from different angles. Imagine the ribosome as a walnut with two nutshell halves of slightly different sizes held together by the seam (in botany: suture).
Rotate the walnut so that the suture runs vertically from the tip (top) to the bottom, where the nut was once attached to the leafstalk (in botany: petiole). The smaller half of the nutshell (30S subunit) is on the left side, and the larger half (50S subunit) is on the right side. Now you're ready to move (leap?) from the walnut to the model of the crystal structure of the E. coli 70S ribosome shown in Figure 3.1. (An aside: you get this view of the ribosome from Figure 2 if you rotate it 90° clockwise.)
Figure 3.1. Side view of the 70S ribosome without mRNA. 30S small subunit (16S RNA cyan, proteins dark magenta); 50S large subunit (23S RNA gray, 5S RNA pale violet, proteins magenta); three bound tRNAs (E site red, P site orange, A site yellow). Image from the Noller Lab website, UC Santa Cruz, USA.
In this model of the 70S ribosome derived from X‑ray crystallography with a resolution of ~2 Å, virtually all atoms are visible except for hydrogen atoms (diameter ~1.1 Å). All nucleotides of the rRNA molecules (16S rRNA cyan, 23S rRNA grey, 5S rRNA pale magenta) can be accurately assigned, by convention shown as helical loops for double-stranded segments and strings for single-stranded segments. Also all ribosomal proteins (magenta) with their amino acid side chains and tertiary structures can be accurately assigned, by convention as loops for α‑helices and arrows for β‑sheets. Note that the ribosomal proteins by no means wrap around the rRNAs like a shell, but rather appear as strategically sprinkled around and within the ribosome.
Unlike a solid, opaque walnut shell, both ribosomal subunits are, in fact, partially "transparent" and allow to peek into the ribosom's "innards" with its gorges and crevices. However, because of the proteins and rRNAs covering them, you can only see small parts of the A site-bound tRNA (yellow), the P site-bound tRNA (orange), and the E site-bound tRNA (red), and you cannot make out their exact positions inside the ribosome. But can you see the positioning of the tRNAs within the ribosome when it is flipped by 180° around the vertical axis, from a "rear view"? In Figure 3.2, you see the 30S subunit on the right and the 50S subunit on the left:
Figure 3.2. Side view of the 70S ribosome without mRNA. 30S small subunit (16S RNA cyan, "S" proteins dark magenta); 50S large subunit (23S RNA gray, 5S RNA pale violet, "L" proteins magenta); three bound tRNAs (E site red, P site orange, A site yellow); LY, LW unidentified proteins in earlier maps. Image from the Noller Lab website, UC Santa Cruz, USA.
Again, the tRNA in the A site (yellow) is not visible at all in the "rear view." As for the tRNA in the P site (orange), a small part of its anticodon loop is visible within the 30S subunit, buried within helices of the 16S RNA. Only the 3'‑CCA tail of the tRNA in the E site (red) is visible through the tangles of 23S RNA helices, partially covered by the L1 protein, while its anticodon loop is hidden within the 16S RNA of the 30S subunit and covered by the S7 and S11 proteins.
Taken together: from the outside, the positions of the 3'-CCA tail of both the tRNA in the A site and the tRNA in the P site within the ribosome – and thus the peptidyltransferase center (PC) – are not visible, they are buried within the "innards" of the ribosome. Enter the nutcracker. If you split the walnut along the suture and rotate each half 90°, that is, flipped to the right for the 50S subunit, and to the left for the 30S subunit, you get a view of the inside: you have the 30S–50S interface right in front of you.
In Figure 3.3, you see the interface of the large 50S subunit, and below in Figure 3.4 that of the small 30S subunit. In both views, the three tRNAs (E red, P orange, A yellow) are shown anchored in their respective subunits, facing toward you, thus appearing in the foreground. This means that in Figure 3.3, you are looking directly at the anticodon loop of the tRNAs, and in Figure 3.4, you are looking directly at their 3'‑CCA tails – their "business end" for the peptidyltransferase reaction.
Figure 3.3. Interface view of the 50S large subunit (23S RNA gray, 5S RNA pale violet, "L" proteins magenta); three bound tRNAs (E site red, P site orange, A site yellow); ASF A‑site finger, SRL sarcin-ricin loop, LY unidentified protein in earlier maps. Image from the Noller Lab website, UC Santa Cruz, USA.
Figure 3.4. Interface view of the 30S large subunit (16S RNA cyan, "S" proteins dark magenta); three bound tRNAs (E site red, P site orange, A site yellow); PS platform stem . Image from the Noller Lab website, UC Santa Cruz, USA.
You see in Figure 3.4 that the 3'-CCA tails of the tRNAs bound to the A site and the P site come very close to each other within the 50S subunit. This is even more apparent in the "open book" view of the ribosome in Figure 3.5.. Note that the 3D-structure of a tRNA does not resemble the familiar cloverleaf structure but more a capital L, with the longer arm respresenting the anticodon stem (contacting the codon triplet in the mRNA), the shorter arm the 3'-CCA tail (buried within the 50S subunit), and the kink of the L, the lush hip. Also note that tRNAs are actually quite big, or bulky, and have Svedberg values of about 4S compared to the 5S, 16S, and 23S rRNAs (and are not these twisted & knotted mini-structures shown in Figure 2). They are also more massive than most ribosomal proteins.
Figure 3.5. "Open book" view of the 70S ribosome (without mRNA) reveals the positions of the three tRNAs (E site red, P site orange, A site yellow) relative to the 30S small subunit (16S RNAcyan, "S" proteins dark blue) and the 50S large subunit (23S RNA gray, 5S RNA pale violet, "L" proteins magenta). Frontispiece: Detail from the "open book" view. Image from the Noller Lab website, UC Santa Cruz, USA.
The model in which three tRNAs bind to the ribosome simultaneously is structurally sound, but a bit of a stretch: in reality, it occurs only for the instant when a newly incoming aminoacyl-tRNA binds into the empty A site and (allosterically) triggers the ejection of the de-charged tRNA from the E site. Think of it this way: The peptidyltransferase reaction also takes place within this very same brief time window. More on this in part 2.
We'd be delighted to hear from you! You can comment below, send us an email or reach us via Bluesky or Mastodon. Please subscribe to get email notifications of our new posts.
comment by Roland Hatzenpichler on BlueSky
Rarely have I met a (undergrad or grad) student who knew about the role of 23S rRNA, it's importance as RNAzyme, and the implications for how we view the origin of RNA and proteins. I don't know why they are not being taught this in their classes. It's one of the most important concepts in biology.
"It" referring to RNA is not only template but also creates protein.
Does this exclusive focus on the ribosomal PTC mean that Substrate-assisted catalysis of peptide bond formation by the tRNA substrate itself has been ruled out by now, or has that view just been marginalized until it fell into oblivion?
to the best of my knowledge, peptide bond formation involving only the pept/aminoacyl tRNAs and not 23S rRNA residues is not considered anymore, since ~2005. Please see part 2.