Braille is a system of raised dots for translating written words into tactile word signs.
(Hulton-Deutsch Collection/Corbis)

Chapter 33

Protein Synthesis and Degradation

Protein biosynthesis is achieved by the process of translation. Translation converts the language of genetic information embodied in the base sequence of a messenger RNA molecule into the amino acid sequence of a polypeptide chain. During translation, proteins are synthesized on ribosomes by linking amino acids together in the specific linear order stipulated by the sequence of codons in an mRNA. Ribosomes are the agents of protein synthesis.

33.1 · Ribosome Structure and Assembly

Ribosomes are compact ribonucleoprotein particles found in the cytosol of all cells, as well as in the matrix of mitochondria and the stroma of chloroplasts. The general structure of ribosomes is described in Chapter 11; here we consider their structure in light of their function in synthesizing proteins. Ribosomes are mechano-chemical systems that move along mRNA templates, orchestrating the interactions between successive codons and the corresponding anticodons presented by aminoacyl-tRNAs. As they align successive amino acids via codon-anticodon recognition, ribosomes also catalyze the formation of peptide bonds between adjacent amino acid residues.

The Composition of Prokaryotic Ribosomes

Escherichia coli ribosomes are representative of the structural organization of the prokaryotic versions of these supramolecular protein-synthesizing machines (Table 33.1, see also Figure 11.25). The E. coli ribosome is a roughly globular particle with a diameter of 25 nm, a sedimentation coefficient of 70S, and a mass of about 2520 kD. It consists of two unequal subunits that dissociate from each other at Mg^{2 1} concentrations below 1 mM. The smaller, or 30S, subunit has a mass of 930 kD and is composed of 21 different proteins and a so-called 16S ribosomal RNA (rRNA) molecule 1542 nucleotides long. The larger 50S subunit has a mass of 1590 kD and consists of 31 different polypeptides (L1 to L34, see later) and two rRNAs, a 2904-nucleotide 23S rRNA and a 120-nucleotide 5S rRNA. Ribosomes are roughly two-thirds RNA and one-third protein by mass. An E. coli cell contains around 20,000 ribosomes, constituting about 20% of the dry cell mass.

Ribosomal Proteins

There is one copy of each ribosomal protein per 70S ribosome, excepting protein L7/L12 (L7 and L12 have identical amino acid sequences and differ only in the degree of N-terminal acetylation). Only one protein is common to both the small and large subunit: S20 = L26. The amino acid sequences of all 52 E. coli ribosomal proteins are known. The largest is S1 (557 residues, 61.2 kD); the smallest is L34 (46 residues, 5.4 kD). The sequences of ribosomal proteins share little similarity. These proteins are typically rich in the cationic amino acids Lys and Arg and have few aromatic amino acid residues, properties appropriate to proteins intended to interact strongly with polyanionic RNAs. Crystal structures obtained thus far for ribosomal proteins reveal a variety of structures. Because RNA adopts a greater range of structures than DNA, RNA-protein recognition motifs are expected to be diverse.

Figure 33.1 · The seven ribosomal RNA operons in E. coli. These operons, or gene clusters, are transcribed to give a precursor RNA that is subsequently cleaved by RNase III and other nucleases, at the sites indicated, to generate 23S, 16S, and 5S rRNA molecules, as well as several tRNAs that are unique to each operon. Numerals to the right of the brackets indicate the number of species of tRNA encoded by each transcript.

rRNAs

The three E. coli rRNA molecules—23S, 16S, and 5S—are derived from a single 30S rRNA precursor transcript that also includes several tRNAs (Figure 33.1). Ribosomal RNAs show extensive potential for intrachain hydrogen bonding and assume secondary structures reminiscent of tRNAs, although substantially more complex (Figures 12.39 and 12.40). Almost half the bases in 16S rRNA are base paired. Relatively small double-helical regions are punctuated by short, single-stranded stretches, generating hairpin configurations that dominate the molecule; four distinct domains (I through IV) can be discerned in the secondary structure. Models of the three-dimensional structure of E. coli rRNAs show that, to a good approximation, these rRNA structures conform to the general shapes of the ribosomal subunits. Figure 33.2 shows the three-dimensional structure of 16S rRNA within the 30S subunit.

Figure 33.2 · The 16S rRNA within the 30S ribosomal subunit. This view is of the “solvent side” of the 30S subunit (the side opposite to the 50S subunit-binding side—see Figure 33.3).(Adapted from Mueller, F., and Brimacombe, R., 1997. A new model for the three-dimensional folding of Escherichia coli 16S ribosomal RNA: I. Fitting the RNA to a 3D electron microscopic map at 20 Å. Journal of Molecular Biology 271:524-544, Figure 2. Figure courtesy of Florian Mueller and Richard Brimacombe, Max-Planck-Institute for Molecular Genetics, Berlin.)

Self-Assembly of Ribosomes

Ribosomal subunit self-assembly is one of the paradigms for the spontaneous formation of supramolecular complexes from their macromolecular components. If the individual proteins and rRNAs composing ribosomal subunits are mixed together in vitro under appropriate conditions of pH and ionic strength, spontaneous self-assembly into functionally competent subunits takes place without the intervention of any additional factors or chaperones. Apparently, the rRNA acts as a scaffold upon which the various ribosomal proteins convene. Ribosomal proteins bind in a specified order.
            Assembly of 30S subunits begins even as the rRNA precursor is being transcribed. The first part of the 16S rRNA to be transcribed, the 5'-region, possesses a cluster of the strongest protein-binding sites.

Ribosomal Architecture

Ribosomal subunits have a characteristic three-dimensional architecture that has been revealed by image reconstructions fromcryoelectron microscopy and X-ray and neutron solution scattering. Such an analysis of the small ribosomal subunit leads to the
Figure 33.3 · A three-dimensional model for the E. coli ribosome as deduced by image reconstruction. The 30S, 50S, and 70S structures are presented in two views that are rotated 90° relative to each other. The small and large ribosomal subunits each have characteristic morphological features, such as protuberances and clefts. The 30S subunit is somewhat elongated and asymmetric and has the dimensions 5.5 x 22 x 22 nm. The 50S subunit is predominantly spheroidal but has three projections; its dimensions are 15 x 20 x 20 nm. A 2.5-nm tunnel passes through the large subunit in the central “valley” region between the subunit’s three protrusions.

model depicted in Figure 33.3. The small subunit features a “head” and a “base” from which a “platform” projects. A “cleft” is defined by the spatial relationship between the head, base, and platform. The large subunit is a globular structure with three

Figure 33.4 · (a) A comparison of their relative sizes indicates that ribosomes are large enough to accommodate two tRNAs simultaneously. (b) Image reconstruction of the E. coli 70S ribosome from three-dimensional cryoelectron microscopy. The view is perpendicular to the 30S:50S subunit interface, with the 30S subunit on the left and the 50S subunit on the right. The path of mRNA and the location of three tRNA molecules within the central cavity are shown. (See Figure 33.6 for discussion of the three tRNA-binding sites within ribosomes.) (Adapted from Frank, J., 1997. The ribosome at higher resolution—The donut takes shape. Current Opinion in Structural Biology 7:266-272, Figure 2. Figure [b] courtesy of Joachim Frank, Wadsworth Center, New York State Department of Health, Albany.)

distinctive projections: a “central protuberance,” the “stalk,” and a winglike ridge. The two subunits associate with each other so that the side of the small subunit nestles into the cleft of the large subunit, with the platform of the small subunit oriented toward the “wing” of the large subunit (Figure 33.3). An extensive cavity between the two ribosomal subunits is large enough to hold two or more tRNAs, as well as several protein factors involved in protein synthesis (Figure 33.4). The small subunit has a channel leading into the cleft, through which the mRNA passes. The cleft is aligned somewhat with a branched tunnel in the large subunit, and it is supposed that the growing peptidyl chain is threaded through this tunnel as protein synthesis proceeds. Even though the ribosomal proteins are arranged peripherally around the rRNAs in ribosomes, rRNA occupies 30 to 40% of the ribosomal subunit surface areas.

Eukaryotic Ribosomes

Eukaryotic cells have ribosomes in their mitochondria (and chloroplasts) as well as in the cytosol. The mitochondrial and chloroplastic ribosomes resemble prokaryotic ribosomes in size, overall organization, structure, and function, a fact reflecting the prokaryotic origins of these organelles. Whereas eukaryotic cytosolic ribosomes retain many of the structural and functional properties of their prokaryotic counterparts, they are larger and considerably more complex. Further, higher eukaryotes have more complex ribosomes than lower eukaryotes. For example, the yeast cytosolic ribosomes have major rRNAs of 3392 (large subunit) and 1799 nucleotides (small subunit); the major rRNAs of mammalian cytosolic ribosomes are 4718 and 1874 nucleotides, respectively. Table 33.2 lists the properties of cytosolic ribosomes in a representative mammal, the rat. Their mass is almost 1.7 times the mass of E. coli ribosomes, and proteins contribute a relatively greater proportion of this mass. Small (40S) subunits have 33 different proteins and large (60S) subunits have 49. Large subunits have three characteristic rRNAs: 28S, 5.8S, and 5S. The sequence of the 5.8S rRNA shows homology to the 5'-end of prokaryotic 23S rRNA, suggesting it may be an evolutionary derivative of it. This 5.8S rRNA forms a secondary structure with 28S rRNA through complementary base pairing. Comparison of base sequences and secondary structures of rRNAs from different organisms suggests that evolution has worked to conserve the secondary structure of these molecules, although not necessarily the nucleotide sequences creating such structure. That is, the retention of a base pair at a particular location seems more important than whether the base pair is G:C or A:U. The morphology of eukaryotic cytosolic ribosomes resembles that of prokaryotic ribosomes.

Conserved Bases in rRNA Are Clustered in Single-Stranded Regions

Although retention of secondary structure apparently plays a leading role in rRNA evolution, conservation of rRNA primary structure at particular places is also a significant feature of these molecules. Comparison of the base sequences of 16S-like (rRNAs) (the small subunit rRNA) from diverse phylogenetic sources reveals that about one-third of the nucleotides are universally conserved. Interestingly, many of these nucleotides are clustered in a few single-stranded regions of the molecule. Several such sequences on the order of 10 to 20 nucleotides long are essentially invariant in all organisms examined. Such conservation strongly suggests that these unpaired stretches play some functional role during protein synthesis.

33.2 · The Mechanics of Protein Synthesis

Like chemical polymerization processes, protein biosynthesis in all cells is characterized by three distinct phases: initiation, elongation, and termination. At each stage, the energy driving the assembly process is provided by GTP hydrolysis, and specific soluble protein factors participate in the events.
            Initiation involves binding of mRNA by the small ribosomal subunit, followed by association of a particular initiator aminoacyl-tRNA that recognizes the first codon. This codon often lies within the first 30 nucleotides or so of mRNA spanned by the small subunit. The large ribosomal subunit then joins the initiation complex, preparing it for the elongation stage.
            Elongation includes the synthesis of all peptide bonds from the first to the last. The ribosome remains associated with the mRNA throughout elongation, moving along it and translating its message into an amino acid sequence. This is accomplished via a repetitive cycle of events in which successive aminoacyl-tRNAs add to the ribosome:mRNA complex as directed by codon binding, and the polypeptide chain grows by one amino acid at a time.

Figure 33.5 · The basic steps in protein synthesis. Note that the ribosome has two principal sites for binding tRNA: the A, or acceptor, site and the P, or peptidyl, site.

            Only two tRNA molecules are part of the ribosome:mRNA complex at any moment. Each lies in a distinct site (Figure 33.5). The A, or acceptor, site is the attachment site for an incoming aminoacyl-tRNA. The P, or peptidyl, site is occupied by peptidyl-tRNA, the tRNA carrying the growing polypeptide chain. The elongation reaction transfers the peptide chain from the peptidyl-tRNA in the P site to the aminoacyl-tRNA in the A site. This transfer occurs through covalent attachment of the peptidyl a-carboxyl to the a-amino of the aminoacyl-tRNA, forming a new peptide bond. The new, longer peptidyl-tRNA now moves from the A site into the P site as the ribosome moves one codon further along the mRNA. The A site, left vacant by this translocation, can accept the next incoming aminoacyl-tRNA. These events are summarized in Figure 33.5. A third tRNA-binding site on the ribosome, the E, or exit, site, is transiently occupied by deacylated tRNAs as they exit the P site, having lost their peptidyl chains. Figure 33.6 depicts plausible locations for the tRNA-binding sites on ribosomes. Elongation is the most rapid phase of protein synthesis.

Figure 33.6 · Plausible locations for tRNA-binding sites within the ribosome. The A, or acceptor site, binds the incoming aa-tRNA, as directed by the mRNA codon underlying it, the pink-shaded area representing the approximate site of interaction of EF-Tu. The P, or peptidyl, site is occupied by the peptidyl-tRNA in association with the mRNA codon previously occupying the A site. The likely path of a tRNA as it transits the events of protein synthesis is shown by the arrows. Juxtaposition of the acceptor stems of the tRNAs in the A and P sites facilitates peptidyl transfer. (Adapted from Noller, H. F., 1991. Ribosomal RNA and translation. Annual Review of Biochemistry 60:191-277.) 

          Termination is triggered when the ribosome reaches a “stop” codon on the mRNA. At this point, the polypeptide chain is released, and the ribosomal subunits dissociate from the mRNA.
            Protein synthesis proceeds rapidly. In vigorously growing bacteria, about 20 amino acid residues are added to a growing polypeptide chain each second. So an average protein molecule of about 300 amino acid residues is synthesized in only 15 seconds. Eukaryotic protein synthesis is only about 10% as fast. We focus first on protein synthesis in E. coli, the most intensively studied system.

Peptide Chain Initiation in Prokaryotes

The components required for peptide chain initiation include (a) mRNA, (b) 30S and 50S ribosomal subunits, (c) a set of proteins known as initiation factors, (d) GTP, and (e) a specific charged tRNA, f-Met-tRNA_f^Met. A discussion of the properties of these components and their interaction follows.

Initiator tRNA

Figure 33.7 · The structure of E. coli N-formyl-methionyl-tRNA_f^Met. The features distinguishing it from noninitiator tRNAs are highlighted.

tRNA_f^Met is a particular tRNA for reading an AUG (or GUG, or even UUG) codon that signals the start site, or N-terminus, of a polypeptide chain. This tRNA_f^Met does not read internal AUG codons, so it does not participate in chain elongation. Instead, that role is filled by another methionine-specific tRNA, referred to as tRNA_m^Met, which cannot replace tRNA_f^Met in peptide chain initiation. (However, both of these tRNAs are loaded with Met by the same methionyl-tRNA synthetase.) The structure of E. coli tRNA_f^Met has several distinguishing features (Figure 33.7). Unlike the case with all other tRNAs, the 5'-terminal base is not matched with a complementary base in the tRNA_f^Met acceptor stem, and thus no base pair forms here. tRNA_f^Met also has a unique CCU sequence in its D loop and an exclusive set of three G:C base pairs in its anticodon stem. Collectively, these features identify this tRNA as essential to initiation and inappropriate for chain elongation.
     The synthesis of all E. coli polypeptides begins with the incorporation of a modified methionine residue, N-formyl-Met, as N-terminal amino acid. However, in about half of the E. coli proteins, this Met residue is removed once the growing polypeptide is 10 or so residues long; as a consequence, many mature proteins in E. coli lack N-terminal Met.
Figure 33.8 · Methionyl-tRNA_f^Met formyl transferase catalyzes the transformylation of methionyl-tRNA_f^Met using N¹⁰-formyl-THF as formyl donor. tRNA_m^Met is not a substrate for this transformylase.

     The methionine contributed in peptide chain initiation by tRNA_f^Met is unique in that its amino group has been formylated. This reaction is catalyzed by a specific enzyme, methionyl-tRNA_f^Met formyl transferase (Figure 33.8). Note that the addition of the formyl group to the Met-NH₂ creates an N-terminal block resembling a peptidyl grouping. That is, the initiating Met is transformed into a minimal analog of a peptidyl chain.

mRNA Recognition and Alignment

Figure 33.9 · Various Shine-Dalgarno sequences recognized by E. coli ribosomes. These sequences lie about 10 nucleotides upstream from their respective AUG initiation codon and are complementary to the UCCU core sequence element of E. coli 16S rRNA. G:U as well as canonical G:C and A:U base pairs are involved here.

In order for the mRNA to be translated accurately, its sequence of codons must be brought into proper register with the translational apparatus. Recognition of translation initiation sequences on mRNAs involves the 16S rRNA component of the 30S ribosomal subunit. Base pairing between a pyrimidine-rich sequence at the 3'-end of 16S rRNA and complementary purine-rich tracts at the 5'-end of prokaryotic mRNAs positions the 30S ribosomal subunit in proper alignment with an initiation codon on the mRNA. The purine-rich mRNA sequence, the ribosome-binding site, is often called the Shine-Dalgarno sequence in honor of its discoverers. Figure 33.9 shows various Shine-Dalgarno sequences found in prokaryotic mRNAs, along with the complementary 3'-tract on E. coli 16S rRNA. The 3'-end of 16S rRNA resides in the “head” region of the 30S small subunit.
     These recognition events are verified by studies of prokaryotic ribosomes treated with the bacteriocidal protein colicin E3. This protein is a phospho-diesterase that specifically cleaves the bond after position 1493 in 16S rRNA, removing the 3'-terminal 49 nucleotides of 16S rRNA that include the pyrimidine-rich Shine-Dalgarno-binding sequence (Figure 12.39). Colicin E3-treated ribosomes are competent in the elongation of polypeptide chains whose synthesis has already been initiated, but these ribosomes can no longer initiate mRNA translation.

Initiation Factors

Initiation involves interaction of the initiation factors (IFs) with GTP, N-formyl-Met-tRNA_f^Met, mRNA, and the 30S subunit to give a 30S initiation complex to which the 50S subunit then adds to form a 70S initiation complex. The initiation factors are soluble proteins required for assembly of proper initiation complexes. Their properties are summarized in Table 33.3. The requirement for these proteins was discovered when it was found that 30S ribosomal subunits “washed” with 1M NH₄Cl were inactive in initiating protein synthesis, unless loosely associated proteins that had been removed in the “wash” were added back.

Events in Initiation

Figure 33.10 · The sequence of events in peptide chain initiation.

Initiation begins when a 30S subunit:(IF-3:IF-1) complex binds mRNA and a complex of IF-2, GTP, and f-Met-tRNA_f^Met. The sequence of events is summarized in Figure 33.10. Although IF-3 is absolutely essential for mRNA binding by the 30S subunit, it is not involved in locating the proper translation initiation site on the message. The presence of IF-3 on 30S subunits also prevents them from reassociating with 50S subunits. IF-3 must dissociate before the 50S subunit will associate with the mRNA:30S subunit complex.
            IF-2 delivers the initiator f-Met-tRNA_f^Met in a GTP-dependent process. Apparently, the 30S subunit is aligned with the mRNA such that the initiation codon is situated within the “30S part” of the P site. Upon binding, f-Met-tRNA_f^Met enters this 30S portion of the P site. The GTP analog, GMPPCP (Figure 33.11), can replace GTP in promoting IF-2-mediated binding of initiator tRNA to mRNA and the 30S subunit. However, GTP hydrolysis is necessary to form an active 70S ribosome. GTP hydrolysis is accompanied by IF-1 and IF-2 release and probably occurs when the 50S subunit joins. It is likely that GTP hydrolysis is catalyzed by a ribosomal protein, not IF-2. In any event, GTP hydrolysis is believed to drive a conformational alteration that renders the 70S ribosome competent in chain elongation. The A site of the 70S initiation complex is poised to accept an incoming aminoacyl-tRNA.

Figure 33.11 · The structure of GMPPCP (also known as GDPCP), a nonhydrolyzable analog of GTP. The advantage of GDPCP is that it allows the events promoted by GTP binding to be separated from those promoted by GTP hydrolysis.

Peptide Chain Elongation

The requirements for peptide chain elongation are (a) an mRNA:70S ribosome:peptidyl-tRNA complex (peptidyl-tRNA in the P site), (b) aminoacyl-tRNAs, (c) a set of proteins known as elongation factors, and (d) GTP. Chain elongation can be divided into three principal steps:

1.  Codon-directed binding of the incoming aminoacyl-tRNA at the A site.

2.  Peptide bond formation: transfer of the peptidyl chain from the tRNA bearing it to the -NH₂ group of the new amino acid.

3.  Translocation of the “one-residue-longer” peptidyl-tRNA to the P site to make room for the next aminoacyl-tRNA at the A site. These shifts are coupled with movement of the ribosome one codon further along the mRNA.

The Elongation Cycle

The properties of the soluble proteins essential to peptide chain elongation are summarized in Table 33.4. These proteins are present in large quantities, reflecting the great importance of protein synthesis to cell vitality. For example, elongation factor Tu (EF-Tu) is the most abundant protein in E. coli, accounting for 5% of total cellular protein.

Aminoacyl-tRNA Binding

EF-Tu binds aminoacyl-tRNA and GTP. There is only one EF-Tu species serving all the different aminoacyl-tRNAs, and aminoacyl-tRNAs will bind to the A site of active 70S ribosomes only in the form of aminoacyl-tRNA:EF-Tu:GTP complexes. Once the aminoacyl-tRNA is situated in the A site, the GTP is hydrolyzed to GDP and P_i, and the EF-Tu molecules are released as EF-Tu:GDP complexes (Figure 33.12). (The nonhydrolyzable GTP analog GMPPCP permits aminoacyl-tRNA:EF-Tu binding, but no EF-Tu release occurs, and elongation is arrested.) EF-Tu does not interact with f-Met-tRNA_f^Met.

Figure 33.12 · The cycle of events in peptide chain elongation on E. coli ribosomes.

            Elongation factor Ts (EF-Ts) promotes the recycling of EF-Tu by mediating the displacement of GDP from EF-Tu and its replacement by GTP. EF-Ts accomplishes its job through entry into a transient complex with EF-Tu. GTP then displaces EF-Ts from EF-Tu (Figure 33.12).

Peptidyl Transfer

Figure 33.13 · Reaction of the tRNA-linked peptidyl chain with the a-amino group of an adjacent aminoacyl-tRNA requires no external energy of activation (as, for example, from ATP).

Peptidyl transfer, or transpeptidation, is the central reaction of protein synthesis, the actual peptide bond-forming step. No energy input (for example, in the form of ATP) is needed; the ester bond linking the peptidyl moiety to tRNA is intrinsically reactive (Figure 33.13). Peptidyl transferase, the activity catalyzing peptide bond formation, is associated with the 50S ribosomal subunit.

Figure 33.14 · The Peptidyl transferase center of 23S rRNA. 23S rRNA has a highly conserved secondary sturcture reminiscent of that of I6S rRNA (see Figure 12.39). Only sequences implicated in the peptidyl trnasferase function are presented here. Numbers indicated bas positions in the 23S rRNA nucleotide sequence. Green dots symbolize bases in the central region that are not conserved in 23SrRNAs from different sources. 23S rRNA sites involved in interactions with peptidyl-tRNA and aminoacyl-tRNA, the substrates of protein synthessi, are indicated. Residue G2252 forms close contacts with the CCA-end of the P-site (peptidyl) tRNA, and residue G2553 forms close contacts with the CCA-end of the A-site (aminoacyl)tRNA. The loops in which these residues are found are labeled P loop and A loop, respectively. The results indicate tha the tertiary sturcuture of 23SrRNA brings the P loop-bound peptidyl-tRNA into an interaction with base U2585 inn the central region. (Adapted from Pace, N.R., 1992. New horizons for RNA catalysis. Science 256:1402-1403, Porse, B., et al., 1997. The donor substrate site within the peptidyl transferase loop of 23S rRNA and its putative interactions with the CCA-end of N-blocked aminoacyl-tRNA. Journal of Molecular Biology 266:472-483, and Green,R.,et al., 1998. Ribosome-catalyzed peptide-bond formation with an A-site substrate covalently linked to 23S ribosomal RNA. Science 280:286-289.)

23S rRNA Is the Peptidyl Transferase Enzyme

Remarkably, E. coli 50S ribosomal subunits from which virtually all ribosomal proteins have been removed retain significant peptidyl transferase activity (as assayed by a simplified model reaction that mimics the usual peptidyl transferase assay, Figure 14.24). These experiments, carried out by Harry Noller and his colleagues, imply that the peptidyltransferase enzyme is the 23S rRNA. The peptidyl transferase center of 23S rRNA is depicted in Figure 33.14. Nucleotide sequences in this region of 23S rRNA are among the most highly conserved in all biology.

Translocation

Three things remain to be accomplished in order to return the active 70S ribosome:mRNA complex to the starting point in the elongation cycle:

1. The deacylated tRNA must be removed from the P site.

2. The peptidyl-tRNA must be moved (translocated) from the A site to the P site.

3. The ribosome must move one codon down the mRNA so that the next codon is positioned in the A site.

Figure 33.15 · Model of the binding states, (a) through (e), for the movement of tRNAs during translation. In this model, the hybrid states are A/T, E/P, and P/A. The numerator represents the hypothetical 50S condition, and the denominator is the corresponding 30S condition. (Adapted from Noller, H. F., 1991. Ribosomal RNA and translation. Annual Review of Biochemistry 60:191-227.)

            The precise events in translocation are still being resolved, but several distinct steps are clear. Within the 70S ribosome, the anticodon ends of both the A-site and P-site tRNAs interact with the 30S subunit within an area referred to as the decoding center; it is here that mRNA codon:tRNA anticodon recognition takes place. In contrast, the acceptor ends (the aminoacylated ends) of both A-site and P-site tRNAs interact with the 50S subunit. In the first step, the acceptor ends of the A- and P-site tRNAs move with respect to the 50S subunit concomitantly with peptidyl transfer (Figure 33.15c-d). Nucleophilic attack of the a-amino group of the aminoacyl-tRNA in the A site on the C-terminal carbonyl carbon of the peptidyl-tRNA results in peptide bond formation (see Figure 33.13) and transfer of the peptidyl chain to the A-site tRNA. Because the growing peptidyl chain does not move during peptidyl transfer, this reaction brings the acceptor end of the A-site tRNA into the P site as it picks up the peptidyl chain. The acceptor end of the P-site tRNA is shunted into the E site. These movements result in two hybrid states of tRNA binding: the E/P state and the P/A state (Figure 33.15d). Thus, the two ends of a tRNA are associated with different sites on the two ribosomal subunits (for example, in the P/A state, the peptidyl chain is linked to a tRNA having its acceptor end in the 50S P site and its anticodon end in the 30S A site). In the second step, the tRNAs and mRNA move together with respect to the 30S subunit so that the mRNA is passively dragged one codon along as the anticodon end of the peptidyl-tRNA is transferred from the A site of the 30S subunit to its P site (Figure 32.15e). Concomitantly, the deacylated tRNA is moved into the E site. These movements are catalyzed by the translocation protein elongation factor G (EF-G), which couples the energy of GTP hydrolysis to movement (see Figures 33.12 and 33.15). Translocation of the mRNA relative to the 30S subunit delivers the next codon to the 30S A site. EF-G binds to the ribosome as an EF-G:GTP complex. GTP hydrolysis is essential not only for translocation but also for subsequent EF-G dissociation. Because EF-G and EF-Tu compete for a common binding site on the ribosome, EF-G release is a prerequisite for return of the 70S ribosome:mRNA to the beginning point in the elongation cycle.

Figure 33.16 · Model showing the relative positions of the two tRNA molecules in a ribosome during the events of peptidyl transfer and translocation. These tRNAs are shown as L-shaped tubes; their relative movements are indicated by dotted outlines. The anticodon ends of the tRNAs are oriented toward the cleft of the small subunit (these ends have circles in the figure). Note that the acceptor ends of the tRNAs are near one another, which facilitates peptide bond formation. Note also the possible role that rotation of the tRNAs about the axis of their anticodon stems and also about the axis of their acceptor stems might have in these events. The mRNA traverses the ribosome along a path that follows the cleft of the 30S subunit. The approximate position of the EF-Tu:aminoacyl-tRNA:GTP/EF-G:GTP binding site is shaded yellow in this figure. (Adapted from Wilson, K., and Noller, H. F., 1998. Molecular movement inside the translational engine. Cell 92:337-349, Figure 4; and Wilson, K., and Noller, H. F., 1998. Mapping the position of EF-G in the ribosome by directed hydroxyl radical probing. Cell 92:131-139, Figure 7.)
            This basic two-step model for translocation (Figure 33.15) identifies six different states of tRNA binding: P/P (peptidyl-tRNA in the P site), A/T (an aminoacyl-tRNA entering the A site), A/A (an aminoacyl-tRNA in the A site), P/A (the peptide chain has been transferred to the aminoacyl-tRNA in the A site), E/P (the deacylated tRNA exiting the P site, and E (the deacylated tRNA in the E site). The A/T state is the first state in tRNA selection at the A site and involves interaction of the ribosome with an aminoacyl-tRNA:EF-Tu:GTP ternary complex. Codon-anticodon recognition and consequent translational fidelity must be determined at this stage. Figure 33.16 illustrates the relative positions of the two tRNA molecules within the 30S:50S intersubunit cavity during these different states.
            In the simple model of peptidyl transfer and translocation, the opposite ends of both tRNAs move relative to the two ribosomal subunits in two discrete steps, the acceptor ends moving first and then the anticodon ends. Further, the readjustments needed to reposition the ribosomal subunits relative to the mRNA and to one another imply that the 30S and 50S subunits must move relative to one another. In proposing that the small and large subunits move relative to each other, as opposed to moving as a unit, this model provides a convincing explanation for why ribosomes are universally organized into a two-subunit structure.

A Deeper Look
Molecular Mimicry — The Structures of EF-Tu:Aminoacyl-tRNA and EF-G
EF-Tu and EF-G compete for binding to ribosomes. EF-Tu has the unique capacity to recognize and bind any aminoacyl-tRNA and deliver it to the ribosome in a GTP-dependent reaction. EF-G catalyzes GTP-dependent translocation. The ternary structure of the Thermus aquaticus EF-Tu:Phe-tRNAPhe complexed with GMPPNP (a nonhydrolyzable analog of GTP) is remarkably similar to the structure of EF-G:GDP (figure). EF-Tu is a 3-domain protein (shown in red, green, and light blue here); a space-filling representation of its bound tRNA is shown in magenta. EF-G is a 6-domain protein, 5 of which correspond to the overall EF-Tu:tRNA structure. Domains 1 and 2 in EF-G correspond to the EF-Tu protein domains colored red and green; domains 3, 4, and 5 of EF-G	show a striking structural resemblance to the tRNA component in EF-Tu:tRNA and are colored magenta. (EF-G has an extra domain, colored dark blue here.) Thus, parts of the EF-G structure mimic the structure of a tRNA molecule.    One view of early evolution suggests that RNA was the primordial macromolecule, fulfilling all biological functions, including those of catalysis and information storage that are now assumed for the most part by proteins and DNA. The mimicry of EF-Tu:tRNA by EF-G may represent a fossil of early macromolecular evolution when the proteins first began to take over some functions of RNA by mimicking shapes known to work as RNAs.
Adapted from Nyborg, J., et al., 1996. Structure of the ternary complex of EF-Tu: Macromolecular mimicry in translation. Trends in Biochemical Sciences 21:81-82.

GTP Hydrolysis Fuels the Conformational Changes That Drive Ribosomal Functions

Note that two GTPs are hydrolyzed for each amino acid residue incorporated into peptide during chain elongation, one upon EF-Tu-mediated binding of aa-tRNA and one more in translocation. The role of GTP (with EF-Tu as well as EF-G) is believed to be mechanical, in analogy with the role of ATP in driving muscle contraction (Chapter 17). GTP binding induces conformational changes in ribosomal components that actively engage these components in the mechanics of protein synthesis; subsequent GTP hydrolysis and GDP and P_i release relax the system back to the initial conformational state so that another turn in the cycle can take place. The energy expenditure for protein synthesis is at least four high-energy phosphoric anhydride bonds per amino acid. In addition to the two provided by GTP, two are expended in amino acid activation via aminoacyl-tRNA synthesis (Figure 32.5).

Figure 33.17 · The events in peptide chain termination.

Peptide Chain Termination

The elongation cycle of polypeptide synthesis continues until the 70S ribosome encounters a “stop” codon. At this point, polypeptidyl-tRNA occupies the P site and the arrival of a “stop” or nonsense codon in the A site signals that the end of the polypeptide chain has been reached (Figure 33.17). These nonsense codons are not “read” by any “terminator tRNAs” but instead are recognized by specific proteins known as release factors, so named because they promote polypeptide release from the ribosome. The release factors bind at the A site. RF-1 (36 kD) recognizes UAA and UAG, while RF-2 (41 kD) recognizes UAA and UGA. There is about one molecule each of RF-1 and RF-2 per 50 ribosomes. Ribosomal binding of RF-1 or RF-2 is competitive with EF-G. The binding of RF-1 or RF-2 is promoted by a third release factor, RF-3 (46 kD). RF-3 function requires GTP.
            The presence of release factors with a nonsense codon in the A site creates a 70S ribosome:RF-1 (or RF-2):RF-3-GTP:termination signal complex that transforms the ribosomal peptidyl transferase into a hydrolase. That is, instead of catalyzing the transfer of the polypeptidyl chain from a polypeptidyl-tRNA to an acceptor aminoacyl-tRNA, the peptidyl transferase hydrolyzes the ester bond linking the polypeptidyl chain to its tRNA carrier. In actuality, peptidyl transferase transfers the polypeptidyl chain to a water molecule instead of an aminoacyl-tRNA. GTP hydrolysis now drives conformational events leading to the dissociation of the uncharged tRNA and expulsion of the release factors from the ribosome (Figure 33.17).

The Ribosome Life Cycle

Figure 33.18 · The ribosome life cycle. Note that IF3 is released prior to 50S addition.

Ribosomal subunits cycle rapidly through protein synthesis. In actively growing bacteria, 80% of the ribosomes are engaged in protein synthesis at any instant. Once a polypeptide chain is synthesized and the nascent polypeptide chain is released, the 70S ribosome dissociates from the mRNA and separates into free 30S and 50S subunits (Figure 33.18). Intact 70S ribosomes are inactive in protein synthesis because only free 30S subunits can interact with the initiation factors. Binding of initiation factor IF-3 by 30S subunits and interaction of 30S subunits with 50S subunits are mutually exclusive. 30S subunits with bound initiation factors associate with mRNA, but 50S subunit addition requires IF-3 release from the 30S subunit.

Figure 33.19 · Electron micrograph of polysomes: multiple ribosomes translating the same mRNA. (From Franke, C., et al., 1982. Electron microscopic visualization of a discrete class of giant translation units in salivary gland cells of Chironomus tentans. The EMBO Journal 1:59-62. Photo courtesy of Oscar L. Miller, University of Virginia.)

Polyribosomes Are the Active Structures of Protein Synthesis

Active protein-synthesizing units consist of an mRNA with several ribosomes attached to it. Such structures are polyribosomes, or, simply, polysomes (Figure 33.19). All protein synthesis occurs on polysomes. In the polysome, each ribosome is traversing the mRNA and independently translating it into polypeptide. The further a ribosome has moved along the mRNA, the greater the length of its associated polypeptide product. In prokaryotes, as many as 10 ribosomes may be found in a polysome. Ultimately, as many as 300 ribosomes may translate an mRNA, so as many as 300 enzyme molecules may be produced from a single transcript. Eukaryotic polysomes typically contain fewer than 10 ribosomes.

The Relationship Between Transcription and Translation in Prokaryotes

Figure 33.20 · The E. coli trp operon. This operon consists of a promoter, p, an operator region, O, and five genes encoding the various enzymes necessary for tryptophan biosynthesis, arranged in the order E, D, C, B, and A. All these genes are transcribed under the control of O and p to give a single polycistronic mRNA.

In prokaryotes, ribosomes attach to mRNA even before transcription of the mRNA is completed, and as a consequence, polysomes can be found in association with DNA. Biochemical evidence for this relationship between transcription and translation comes from a study on the expression of the enzymes for tryptophan biosynthesis in E. coli. These enzymes are encoded in the trp operon, a set of five contiguous genes (Figure 33.20). Transcription of the trp operon occurs only in the absence of tryptophan, and transcription of the entire operon takes more than 5 minutes. The first two genes, E and D, encode the subunits of the enzyme, anthranilate synthase. Within 2 1/2 minutes of the start of trp operon transcription, anthranilate synthase activity is detectable, demonstrating that translation has begun before transcription is completed. That is, the mRNA is translated while the operon is still being transcribed. Oscar Miller, Barbara Hamkalo, and Charles Thomas provided visual evidence for concomitant transcription and translation, as shown in Figure 33.21. The leading ribosome may actually make physical contact with RNA polymerase.

Figure 33.21 · Electron micrograph demonstrating that translation occurs concomitantly with transcription. RNA polymerase initiates transcription (bottom) and moves along the DNA, transcribing it into a complementary RNA strand. Successive ribosomes (dark bodies) attach to the 5'-end of the RNA and begin translating it before transcription is complete, as evidenced by the progressively longer RNA chains with more and more associated ribosomes (top), as RNA polymerase progresses along the DNA strand from the initiation site. (Adapted from Miller, O. L., Jr., Hamkalo, B., and Thomas, C., 1970. Visualization of bacterial genes in action. Science 169:392-395. Photo courtesy of Oscar L. Miller, University of Virginia.)

33.3 · Protein Synthesis in Eukaryotic Cells

Figure 33.22 · The characteristic structure of eukaryotic mRNAs. Untranslated regions ranging between 40 and 150 bases in length occur at both the 5'- and 3'-ends of the mature mRNA. An initiation codon at the 5'-end, invariably AUG, signals the translation start site.

Eukaryotic mRNAs are characterized by two post-transcriptional modifications: the 5'-⁷methyl-GTP capand the poly(A) tail (Figure 33.22). The ⁷methyl-GTP cap is essential for ribosomal binding of mRNAs in eukaryotes and also enhances the stability of these mRNAs by preventing their degradation by 5'-exonucleases. The poly(A) tail enhances both the stability and translational efficiency of eukaryotic mRNAs. The Shine-Dalgarno sequences found at the 5'-end of prokaryotic mRNAs are absent in eukaryotic mRNAs.

Figure 33.23 · The three stages in the initiation of translation in eukaryotic cells. See Table 33.5 for a description of the functions of the eukaryotic initiation factors (eIFs). (Adapted from Pain, V. M., 1996. Initiation of protein synthesis in eukaryotic cells. European Journal of Biochemistry 236:747-771, Figure 1.)

Peptide Chain Initiation in Eukaryotes

The events in eukaryotic peptide chain initiation are summarized in Figure 33.23, and the properties of eukaryotic initiation factors, symbolized as eIFs, are presented in Table 33.5. The eukaryotic initiator tRNA is a unique tRNA functioning only in initiation. Like the prokaryotic initiator tRNA, the eukaryotic version carries only Met. However, unlike prokaryotic f-Met-tRNA_f^Met, the Met on this tRNA is not formylated. Hence, the eukaryotic initiator tRNA is usually designated tRNA_i^Met, with the “i” indicating “initiation.”
            Eukaryotic initiation can be divided into three fundamental steps. Step 1: Association of Met-tRNA_i^Met and initiation factors eIF2, eIF1A, and eIF3 with the 40S ribosomal subunit to form the 43S preinitiation complex (Figure 33.23, stage 1). Met-tRNA_i^Met is delivered to the 40S subunit as an eIF2:GTP:Met-tRNA_i^Met ternary complex. Binding of eIF4A and eIF3 to 40S subunits generates a 43S ribosomal subunit:initiation factor complex; eIF4A catalyzes the association of eIF2:GTP:Met-tRNA_i^Met with the 43S complex to yield the 43S preinitiation complex. Unlike in prokaryotes, binding of Met-tRNA_i^Met by eukaryotic ribosomes occurs in the absence of mRNA, so Met-tRNA_i^Met binding is not codon-directed. Step 2: Binding of the 43S preinitiation complex to mRNA and migration of the 40S ribosomal subunit to the correct AUG initiation codon (Figure 33.23, stage 2). The 43S preinitiation complex binds mRNA at its 5'-terminal ⁷methyl-GTP cap. eIF4E, the mRNA cap-binding protein, represents a key regulatory element in eukaryotic translation. For eIF4E to be active, it must associate with eIF4G to form a complex designated eIF4F. eIF4F also contains eIF4A; eIF4F binding to the cap structure is a prerequisite for association of eIF4B and formation of the 48S preinitiation complex. Translation is inhibited when eIF4E binds with 4E-BP (the eIF4E binding protein). Growth factors stimulate protein synthesis by causing the phosphorylation of 4E-BP, which prevents its binding to eIF4E. The 59-terminal ⁷methyl-GTP cap and the 39-poly(A) tail act synergistically to increase translational effi-ciency. Pab1p, the poly(A)- binding protein, has two binding sites, one for binding to the poly(A) tract on mRNAs and a second for interaction with eIF4G (Figure 33.24). Thus, eIF4G serves as a bridge between the cap-binding eIF4E, the poly(A) tail, and the 40S subunit via eIF3. These interactions initiate scanning of the 40S subunit in search of an AUG codon. Step 3: Addition of the 60S ribosomal subunit to the 48S preinitiation complex, forming the 80S initiation complex, whereupon translation commences (Figure 33.23, stage 3). When the 43S preinitiation complex stops at an AUG codon, GTP hydrolysis in the eIF2:Met-tRNA_i^Met ternary complex causes ejection of the initiation factors bound to the 40S ribosomal subunit. Release of these factors allows 60S subunit association. eIF2:GDP is recycled to eIF2:GTP by eIF2B; eIF2B is a guanine nucleotide exchange factor.

Figure 33.24 · Initiation factor eIF4G serves as a multipurpose adapter to engage the ⁷methyl-G cap:eIF4E complex, the Pab1p:poly(A) tract, and the 40S ribosomal subunit in eukaryotic translation initiation. (Adapted from Hentze, M. W., 1997. eIF4G: A multipurpose ribosome adapter? Science 275:500-501.)

Regulation of Eukaryotic Peptide Chain Initiation

Figure 33.25 · Control of eIF2 functions through reversible phosphorylation of a Ser residue on its a-subunit. The phosphorylated form of eIF2 (eIF2-P) enters a tight complex with eIF2B and is unavailable for initiation.

Regulation of gene expression can be exerted post-transcriptionally through control of mRNA translation. Phosphorylation/dephosphorylation of translational components is a dominant mechanism for control of protein synthesis. Peptide chain initiation, the initial phase of the synthetic process, is the optimal place for such control. Phosphorylation of 40S ribosomal protein S6 facilitates initiation of protein synthesis, resulting in a shift of the ribosomal population from inactive ribosomes to actively translating polysomes. S6 phosphorylation is stimulated by serum growth factors (Chapter 34). The action of eIF4F, the mRNA cap-binding complex, is promoted by phosphorylation. On the other hand, phosphorylation of other translational components inhibits protein synthesis. For example, the a-subunit of eIF2 can be reversibly phosphorylated at a specific Ser residue by an eIF2a kinase/phosphatase system (Figure 33.25). Phosphorylation of eIF2a ultimately inhibits peptide chain initiation, not because eIF2a-P is ineffectual, but because phosphorylated eIF2 binds eIF2B much more tightly than does eIF2. All of the eIF2B, which is present at only 20 to 30% of eIF2 levels, becomes sequestered in eIF2:eIF2B complexes, and eIF2 cannot be regenerated from eIF2:GDP for further cycles of initiation. Reversible phosphorylation of eIF2 is an important control governing globin synthesis in reticulocytes. If heme for hemoglobin synthesis becomes limiting in these cells, eIF2a is phosphorylated so globin mRNA is not translated and chains are not synthesized. Availability of heme reverses the inhibition through phosphatase-mediated removal of the phosphate group from the Ser residue.

Peptide Chain Elongation in Eukaryotes

Eukaryotic peptide elongation occurs in very similar fashion to the process in prokaryotes. An incoming aminoacyl-tRNA enters the ribosomal A site while peptidyl-tRNA occupies the P site. Peptidyl transfer then occurs, followed by translocation of the ribosome one codon further along the mRNA. Two elongation factors, EF1 and EF2, mediate the elongation steps. EF1 consists of two components: EF1A, a 50-kD protein, and EF1B, a complex of 31-kD (b) and 50-kD (g) protein subunits, EF1A is the eukaryotic counterpart of EF-Tu; it serves as the aminoacyl-tRNA binding factor and requires GTP. EF1B is the eukaryotic equivalent of prokaryotic EF-Ts; it catalyzes the exchange of bound GDP on EF1:GDP for GTP so active EF1:GTP can be regenerated. EF2, a 100-kD polypeptide, is the eukaryotic translocation factor. Like its prokaryotic kin, EF-G, EF2 binds GTP, and GTP hydrolysis accompanies translocation.

Eukaryotic Peptide Chain Termination

Whereas prokaryotic termination involves three different release factors (RFs), just one RF is sufficient for eukaryotic termination. Eukaryotic RF (110 kD) is an a₂ dimer of 55-kD subunits. Eukaryotic RF binding to the ribosomal A site is GTP-dependent, and RF:GTP binds at this site when it is occupied by a termination codon. Then, hydrolysis of the peptidyl-tRNA ester bond, hydrolysis of GTP, release of nascent polypeptide and deacylated tRNA, and ribosome dissociation from mRNA ensue.

Figure 33.26 · The structures of various anti-biotics that act as protein synthesis inhibitors. Puromycin mimics the structure of aminoacyl-tRNA in that it resembles the 39-terminus of a Tyr-tRNA.

33.4 · Inhibitors of Protein Synthesis

Protein synthesis inhibitors have served two major, and perhaps complementary, purposes. First, they have been very useful scientifically in elucidating the biochemical mechanisms of protein synthesis. Second, some of these inhibitors affect prokaryotic but not eukaryotic protein synthesis and thus are medically important antibiotics. Table 33.6 is a partial list of these inhibitors and their mode of action. The structures of some of these compounds are given in Figure 33.26.

Streptomycin

Streptomycin is an aminoglycoside antibiotic that affects the function of the prokaryotic 30S subunit, as demonstrated by the fact that streptomycin-resistant mutations map either to the gene encoding 30S protein S12 or to position 912 in the 16S rRNA sequence. Low concentrations of streptomycin induce mRNA misreading, so that improper amino acids are incorporated into the polypeptide. Codons with pyrimidines in the first and second positions are particularly susceptible to streptomycin-induced misreading. These reading errors are not frameshift mistakes, so totally aberrant proteins are not made at low streptomycin levels. Thus, susceptible cells are not killed, but their growth rate is severely depressed. When streptomycin is present at high concentrations, nonproductive 70S ribosome:mRNA complexes accumulate, preventing the formation of active initiation complexes with new mRNA.

Puromycin

Puromycin is a structural analog of the aminoacyl-adenylyl grouping characteristic of the 3'-end of aminoacyl-tRNAs (Figure 33.26). Puromycin binds at the A site of both prokaryotic and eukaryotic ribosomes. Puromycin binding is not dependent on EF-Tu (or EF1). Puromycin serves as an acceptor of the peptidyl chain from peptidyl-tRNA in the P site, in a reaction in which peptidyl transferase catalyzes the attachment of the peptidyl chain to the free NH₃⁺ group of puromycin. Peptidyl-puromycin is a dead-end product because the peptidyl chain is now linked via an amide bond to the 3'-NH of the modified adenosine moiety, not via the usual ester bond to the 3'-OH terminus found in all tRNAs. Puromycin aborts protein synthesis through premature termination, leading to the release of nonfunctional, truncated polypeptides.

Figure 33.27 · Diphtheria toxin catalyzes the NAD ¹-dependent ADP-ribosylation of selected proteins. ADP-ribosylation of the diphthamide moiety of eukaryotic EF2. (Diphthamide=2-[3-carboxamido-3-(trimeth-ylammonio)propyl]histidine.)

Diphtheria Toxin

Diphtheria arises from infection by Corynebacterium diphtheriae bacteria carrying bacteriophage corynephage b. Diphtheria toxin is a phage-encoded enzyme secreted by these bacteria that is capable of inactivating a number of GTP-dependent enzymes through covalent attachment of an ADP-ribosyl moiety derived from NAD⁺. That is, diphtheria toxin is an NAD⁺-dependent ADP-ribosylase. One target of diphtheria toxin is the eukaryotic translocation factor, EF2. This protein has a modified His residue known as diphthamide. Diphthamide is generated post-translationally on EF2; its biological function is unknown. (EF-G of prokaryotes lacks this unusual modification and is not susceptible to diphtheria toxin.) Diphtheria toxin specifically ADP-ribosylates an imidazole-N within the diphthamide moiety of EF2 (Figure 33.27). ADP-ribosylated EF2 retains the ability to bind GTP but is unable to function in protein synthesis. Because diphtheria toxin is an enzyme and can act catalytically to modify many molecules of its target protein, just a few micrograms suffice to cause death.

Ricin

Ricin is an extremely toxic glycoprotein produced by the plant Ricinus communis (castor bean). The protein is a disulfide-linked, ab heterodimer of roughly equal 30-kD subunits. The A subunit (32 kD) is an enzyme and serves as the toxic subunit; it gains entry to cells because the B subunit (33 kD) is a lectin. (Lectins form a class of proteins that bind to specific carbohydrate moieties commonly displayed by glycoproteins and glycolipids on cell surfaces.) Endocytosis of bound ricin followed by disulfide reduction releases the A chain, which gains access to the cytosol and there catalytically inactivates eukaryotic large ribosomal subunits. A single molecule of ricin A chain in the cytosol can inactivate 50,000 ribosomes and kill a eukaryotic cell! Ricin A chain specifically attacks a single, highly conserved adenosine (an A at position 4256) in the eukaryotic 28S rRNA, through an N-glycosidase activity that removes the adenine base, leaving the rRNA sugar-phosphate backbone intact. Removal of this single base is sufficient to inactivate a 60S large subunit. The adenine in this highly conserved region of the 28S rRNA sequence is believed to be crucial to functions of the 60S subunit that involve EF1 and EF2.

33.5 · Protein Folding

Figure 33.28 · Protein folding pathways. (a) Chaperone-independent folding. The protein folds as it is synthesized on the ribosome (green) (or shortly thereafter). (b) Hsp70-assisted protein folding. Hsp70 (gray) binds to nascent polypeptide chains as they are synthesized and assists their folding. (c) Folding assisted by Hsp70 and chaperonin complexes. The chaperonin complex in E. coli is GroES-GroEL. The chaperonin complex in eukaryotic cells is known as TRiC (for TCP-1 ring complex) or CCT (cytosolic chaperonin containing TCP-1). The majority of proteins fold by pathways (a) or (b). (Adapted from Netzer, W. J., and Hartl, F. U., 1998. Protein folding in the cytosol: Chaperonin-dependent and -independent mechanisms. Trends in Biochemical Sciences 23:68-73, Figure 2.)

The information for folding each protein into its unique three-dimensional architecture resides within its amino acid sequence (primary structure). Proteins begin to fold even as they are being synthesized on ribosomes (Figure 33.28a). Nevertheless, proteins may be assisted in folding by a family of helper proteins known as molecular chaperones (see Chapter 6). Chaperones also serve to shepherd proteins to their ultimate cellular destinations.
            The principal chaperones are the Hsp70 and Hsp60 classes of proteins. In Hsp70-assisted folding, proteins of the Hsp70 class bind to nascent polypeptide chains while they are still on ribosomes (Figure 33.28b). Hsp70 (known as DnaK in E. coli) recognizes exposed, extended regions of polypeptides that are rich in hydrophobic residues. By binding to these regions, Hsp70 prevents nonproductive associations and keeps the polypeptide in an unfolded (or partially folded) state until productive folding interactions can occur. Completion of folding requires release of the protein from Hsp70; release is energy-dependent and is driven by ATP hydrolysis.
            A limited number of proteins requires the Hsp60 class of chaperones,

Figure 33.29 · Structure and function of the GroEL-GroES complex. (a) Space-filling representation and overall dimensions of GroEL-GroES (top view, left; side view, right). GroES is gold; the top GroEL ring is green and the bottom GroEL ring is red. (b) Section through the center of the complex to reveal the central cavity. The GroEL-GroES structure is shown as a Ca carbon trace. ADP molecules bound to GroELare shown as space-filling models. (c) Successive cycles of protein binding and release with the chaperonin central cavity, ATP-dependent release, and progressive folding ends in release of the protein from the complex in its fully folded form. (Figure parts [a] and [b] adapted from Figure 1 in Xu, Z., Horwich, A. L., and Sigler, P. B. [1997] Nature 388:741-7749. Molecular graphics courtesy of Paul B. Sigler, Yale University.)

which are known as chaperonins, for the completion of folding. Chaperonins sequester partially folded molecules from one another (and from extraneous interactions) but still allow folding to proceed. The principal chaperonin in E. coli is the GroES-GroEL complex. GroEL is made of two stacked 7-membered rings of 60-kD subunits that form a cylindrical a₁₄ oligomer 15 nm high and 14 nm wide (Figure 33.29). GroEL has a 5-nm central cavity that is the site of ATP-dependent protein folding. GroES consists of a single 7-membered ring of 10-kD subunits that sits like a dome on GroEL (Figure 33.29). (The eukaryotic analog of GroEL, TRiC, is an 8- or 9-membered double-ring structure of 55-kD subunits. TRiC lacks a GroES counterpart.) A partially folded protein molecule is bound within the central cavity of the 14-subunit complex, where its folding is facilitated in an ATP-dependent fashion. Many cycles of binding of the folding protein to the surface of the central cavity, ATP hydrolysis, release of the protein, and rebinding to the surface take place, with folding steps occurring in the brief intervals when the protein is free from the cavity surface (Figure 33.29c). The folding of rhodanese, a 33-kD protein, requires the hydrolysis of about 130 equivalents of ATP. Once the protein has achieved its fully folded state, it is released from GroEL.

33.6 · Post-Translational Processing of Proteins

Aside from these folding events, release of the completed polypeptide from the ribosome is not necessarily the final step in the covalent construction of a protein. Many proteins must undergo covalent alterations before they become functional. In the course of these modifications, the primary structure of a protein may be altered and/or novel derivations may be introduced into its amino acid side chains. Hundreds of different amino acid variations have been described in proteins, virtually all arising post-translationally. The modified His residue, or diphthamide, of EF2 described earlier is one example, as is its ADP-ribosylation by diphtheria toxin. The list of such modifications is very large; some are rather commonplace, whereas others are peculiar to a single protein. A survey of some of the more prominent chemical groups conjugated to proteins, such as carbohydrates and phosphates, is given in Chapter 5.

Proteolytic Cleavage of the Polypeptide Chain

Proteolytic cleavage, as the most prevalent form of protein post-translational modification, merits special attention. The very occurrence of proteolysis as a processing mechanism seems strange: Why join a number of amino acids in sequence and then eliminate some of them? Three reasons can be cited. First, diversity can be introduced where none exists. For example, a simple form of proteolysis, enzymatic removal of N-terminal Met residues, occurs in many proteins. Met-aminopeptidase, by removing the invariant Met initiating all polypeptide chains, introduces diversity at N-termini. Second, proteolysis serves as an activation mechanism so that expression of the biological activity of a protein can be delayed until appropriate. A number of metabolically active proteins, including digestive enzymes and hormones, are synthesized as larger inactive precursors termed pro-proteins that are activated through proteolysis (see zymogens, Chapter 15). The N-terminal pro-sequence on such proteins may act as an intramolecular chaperone to ensure correct folding of the active site. Third, proteolysis is involved in the targeting of proteins to their proper destinations in the cell, a process known as protein translocation.

Protein Translocation

Proteins targeted for service in membranous organelles or for export from the cell are synthesized in precursor form carrying an N-terminal stretch of amino acid residues, or leader peptide,that serves as a signal sequence. In effect, signal sequences serve as “zip codes” for sorting and dispatching proteins to their proper compartments. Thus, the information specifying the correct cellular localization of a protein is found within its structural gene. Once the protein is routed to its destination, the signal sequence is proteolytically clipped from the protein.
            In addition to the bacterial plasma membrane, a number of eukaryotic membranes are competent in protein translocation, including the membranes of the endoplasmic reticulum (ER), nucleus, mitochondria, chloroplasts, and peroxisomes. Several common features characterize protein translocation systems:

1. Proteins to be translocated are made as preproteins containing contiguous blocks of amino acid sequence that act as sorting signals.

2. Membranes involved in translocation have specific protein receptors exposed on their cytosolic faces.

3. Translocons, complex structures consisting of several proteins with different functions, catalyze movement of the proteins across the membrane, and metabolic energy in the form of ATP, GTP, or a membrane potential is essential.

4. Preproteins are maintained in a loosely folded, translocation-competent conformation through interaction with molecular chaperones.

Prokaryotic Protein Translocation

Figure 33.30 · General features of the N-terminal signal sequences on E. coli proteins destined for translocation: a basic N-terminal region, a central apolar domain, and a nonhelical C-terminal region.

Gram-negative bacteria typically have four compartments: cytoplasm, plasma (or inner) membrane, periplasmic space (or periplasm), and outer membrane. Most proteins destined for any location other than the cytoplasm are synthesized with amino-terminal leader sequences 16 to 26 amino acid residues long. These leader sequences, or signal sequences, consist of a basic N-terminal region, a central domain of 7 to 13 hydrophobic residues, and a nonhelical C-terminal region (Figure 33.30). The conserved features of the last part of the leader, the C-terminal region, include a helix-breaking Gly or Pro residue and amino acids with small side chains located one and three residues before the proteolytic cleavage site. Unlike the basic N-terminal and nonpolar central regions, the C-terminal features are not essential for translocation but instead serve as recognition signals for the leader peptidase, which removes the leader sequence. The exact amino acid sequence of the leader peptide is unimportant. Nonpolar residues in the center and a few Lys residues at the amino terminus are sufficient for successful translocation. The functions of leader peptides are to retard the folding of the preprotein so that molecular chaperones have a chance to interact with it and to provide recognition signals for the translocation machinery and leader peptidase.

Eukaryotic Protein Sorting and Translocation

Eukaryotic cells are characterized by many membrane-bounded compartments. In general, signal sequences targeting proteins to their appropriate compartments are located at the N-terminus as cleavable presequences, although many proteins have internal, noncleaved targeting sequences. Proteolytic removal of the leader sequences is also catalyzed by specialized proteases, but removal is not essential to translocation. No sequence similarity is found among the targeting signals for each compartment. Thus, the targeting information resides in more generalized features of the leader sequences such as charge distribution, relative polarity, and secondary structure.

Figure 33.31 · Synthesis of a eukaryotic integral membrane protein and its translocation via the endoplasmic reticulum.

The Synthesis of Secretory Proteins and Many Membrane Proteins Is Coupled to Translocation Across the ER Membrane

The signals recognized by the endoplasmic reticulum translocation system are virtually indistinguishable from bacterial signal sequences; indeed, the two are interchangeable in vitro. In higher eukaryotes, translation and translocation of many proteins destined for processing via the ER are tightly coupled. That is, translocation across the ER occurs as the protein is being translated on the ribosome. As the N-terminal signal sequence of a preprotein undergoing synthesis emerges from the ribosome, it is detected by a so-called signal recognition particle (SRP; Figure 33.31). SRP is a 325-kD nucleoprotein assembly that contains six polypeptides and a 300-nucleotide 7S RNA. SRP binding of the signal sequence halts further protein synthesis on the ribosome. This prevents release of the growing protein into the cytosol before it reaches the ER and its destined translocation. The SRP-ribosome complex then diffuses to the cytosolic face of the ER, where it binds to the docking protein (also known as the SRP receptor), a heterodimeric protein. The a-subunit is anchored to the membrane by the transmembrane b-subunit; both subunits have GTPase activity. Docking is followed by dissociation of SRP in a GTP-dependent process. After targeting to the membrane, the ribosome resumes protein synthesis, delivering its growing polypeptide to the translocon. The translocon is a complex, multifunctional entity that includes, among other proteins, the SRP receptor and Sec61p, a heterotrimeric complex of membrane proteins (Sec61p is known as SecYEGp in prokaryotes). The a-subunit of Sec61p has 10 membrane-spanning segments, whereas the b- and g-subunits are single TMS proteins. Sec61p serves as the transmembrane channel through which the nascent polypeptide is transported into the ER lumen (Figure 33.31). The pore size of Sec61p is about 2 nm.
            Soon after it enters the ER lumen, the signal peptide is clipped off by membrane-bound signal peptidase. Other modifying enzymes within the lumen introduce additional post-translational alterations into the polypeptide, such as glycosylation with specific carbohydrate residues. ER-processed proteins destined for secretion from the cell or inclusion in vesicles such as lysosomes end up contained within the soluble phase of the ER lumen, but polypeptides destined to become membrane proteins carry 20-residue hydrophobic stop-transfer sequences within their mature domains that arrest their passage across the ER membrane. These proteins remain embedded in the ER membrane with their C-termini on the cytosolic face of the ER. Such membrane proteins arrive at their intended destinations via subsequent processing of the ER.

Mitochondrial Protein Import

Most mitochondrial proteins are encoded by the nuclear genome and synthesized on cytosolic ribosomes. Mitochondria consist of three principal subcompartments: the outer membrane, the inner membrane, and the matrix. Thus, mitochondrial proteins must not only find mitochondria, they must gain access to the proper subcompartment, and once there they must attain a functionally active conformation. In principle, comparable considerations apply to protein import to chloroplasts, organelles with four principal subcompartments (outer membrane, inner/thylakoid membrane, stroma, and thylakoid lumen; Chapter 22).

Figure 33.32 · Structure of an amphiphilic a-helix having basic (+) residues on one side and uncharged and hydrophobic (R) residues on the other.

                Signal sequences on nuclear-encoded proteins destined for the mitochondria are N-terminal cleavable presequences 10 to 70 residues long. These mitochondrial presequences lack contiguous hydrophobic regions. Instead, they have positively charged and hydroxyl-amino acid residues spread along their entire length. These sequences form amphiphilic a-helices (Figure 33.32) with basic residues on one side of the helix and uncharged and hydrophobic residues on the other; that is, mitochondrial presequences are positively charged am