• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. May 18, 2010; 107(20): 9111–9116.
Published online May 3, 2010. doi:  10.1073/pnas.0914300107
PMCID: PMC2889043

Cotranslational structure acquisition of nascent polypeptides monitored by NMR spectroscopy


The folding of proteins in living cells may start during their synthesis when the polypeptides emerge gradually at the ribosomal exit tunnel. However, our current understanding of cotranslational folding processes at the atomic level is limited. We employed NMR spectroscopy to monitor the conformation of the SH3 domain from α-spectrin at sequential stages of elongation via in vivo ribosome-arrested 15N,13C-labeled nascent polypeptides. These nascent chains exposed either the entire SH3 domain or C-terminally truncated segments thereof, thus providing snapshots of the translation process. We show that nascent SH3 polypeptides remain unstructured during elongation but fold into a compact, native-like β-sheet assembly when the entire sequence information is available. Moreover, the ribosome neither imposes major conformational constraints nor significantly interacts with exposed unfolded nascent SH3 domain moieties. Our data provide evidence for a domainwise folding of the SH3 domain on ribosomes without significant population of folding intermediates. The domain follows a thermodynamically favorable pathway in which sequential folding units are stabilized, thus avoiding kinetic traps during the process of cotranslational folding.

Keywords: nascent chains, protein folding, ribosome

Protein biosynthesis is achieved by the conversion of genetic information into linear sequences of amino acids and the subsequent folding of the polypeptide chains into their three-dimensional structures. In vivo, the cotranslational folding of newly synthesized proteins is limited by the kinetics of translation and thus must be tightly coordinated with their synthesis on ribosomes. For some proteins, their cotranslational folding might be delayed until the complete sequence information of a folding unit or domain is available, whereas others may start to acquire structural elements even when their folding information is incomplete and concurrent with their elongation at about 15–20 or 5–7 amino acids per second on bacterial or eukaryotic ribosomes, respectively (1). Several pioneering structural studies on ribosomes have been performed that provide detailed insights into the protein translation machinery (24). The diameter of the ribosomal tunnel could allow for a peptide to adopt α-helical structure, and a recent study suggests that nascent chains can make distinct contacts within the tunnel interior (5). Furthermore, there is evidence that some proteins start to acquire secondary and tertiary structures as soon as they emerge from the ribosomal exit tunnel (6, 7). While recent studies have provided evidence for the structural ordering of nascent chains (810), the folding pathway of elongating polypeptides remains largely unexplored. Moreover, it is unclear whether the tethering of the polypeptide to the translation machinery or intermolecular interactions between the nascent polypeptide and the ribosomal surface play a role in the cotranslational folding process.

In this study, we provide a detailed structural description of the cotranslational folding of a nascent polypeptide by monitoring the structure acquisition of the SH3 domain from α-spectrin on ribosomes. To accomplish this, we produced 15N,13C-labeled SH3 nascent chains of different length in Escherichia coli cells under physiological conditions and subsequently employed high-resolution NMR techniques on purified samples to monitor the folding status at specific stages of peptide elongation. This strategy allowed us to probe in detail the conformation of gradually emerging nascent chains.


Design and Purification of Ribosome-Nascent Chain Complexes.

To investigate the conformation and dynamics of nascent polypeptides during protein synthesis by NMR spectroscopy, we created a set of nascent chains that were arrested at specific points during elongation to provide snapshots of ongoing translation. A prerequisite for such structural investigations of nascent polypeptides bound to translating ribosomes is the generation of highly homogeneous populations of 13C,15N-labeled ribosome-nascent chains (RNCs) in large quantities. Therefore, we designed a vector allowing us to express and label target polypeptides in E. coli, to stall their translation at a desired length and to efficiently purify large quantities of RNCs.

The plasmid-based construct used to generate RNCs (Fig. 1A) consisted of an N-terminal triple StrepII-tag for affinity purification followed by a multiple cloning site (MCS) to introduce target genes encoding the nascent polypeptides. In order to remove the StrepII-tag from nascent polypeptides after purification, we inserted an Smt3-domain in between. Smt3 is specifically recognized by the Ulp1 protease that cleaves the polypeptide downstream of Smt3 and thereby produces nascent chains with authentic N-termini. After the MCS, we placed a recognition site for the TEV protease to release the nascent peptides from the ribosomes on demand. This enabled us to record NMR spectra from polypeptides in the ribosome-bound and unbound state. The C-terminal end of the construct was fused via a linker to a stalling sequence of 27 amino acids (aa) derived from the E. coli SecM. The SecM ribosome-stalling peptide includes the motif An external file that holds a picture, illustration, etc.
Object name is pnas.0914300107eq1.jpg that tightly interacts with the interior of the ribosomal tunnel upon its synthesis and arrests translation of any desired target protein fused to it (11). This strategy made it possible to arrest protein elongation at a specific point. Moreover, the SecM stalling peptide spans the ribosomal tunnel and thus ensures the exposure of the nascent chain on the ribosomal surface as shown previously (12).

Fig. 1.
Schematic representation of arrested RNCs and other constructs used for NMR analysis. (A) The plasmid for production of RNCs consists of an N-terminal triple StrepII tag (Red) followed by a Smt3 domain (Green), a multiple cloning site (MCS), a recognition ...

As model nascent chains we chose the Src-homology 3 (SH3) domain of α-spectrin and its mutant variant SH3-m10 (Fig. 1B). The SH3 domain has been intensely studied in the past, and it is known that wild type SH3 (63 aa, 7 kDa) folds into a compact, orthogonal β-sandwich structure composed of five antiparallel β-strands, whereas SH3-m10 harbors two point mutations (W41G and W42V) that prevent native folding and generate a random coil-like entity (13). Moreover, proteinase K digestion experiments suggested that SH3 but not SH3-m10 has the capacity of structure formation while still bound to ribosomes (12). Thus, this pair of nascent chains, with SH3-m10 as a reference for unfolded nascent SH3, was well suited to study cotranslational protein folding at a molecular level by NMR. To mimic distinct stages during ongoing translation, we additionally fused two truncated SH3 versions (SH3-T1 and SH3-T2), lacking 22 and 5 C-terminal residues of SH3, respectively, to the SecM stalling peptide (Fig. 1B). The stalling peptide had a total length of 31 aa (27 aa derived from SecM and four additional residues), which has been shown previously to guarantee the exposure of nascent polypeptides outside of the ribosome (12, 14).

Importantly, all constructs were produced in vivo and thus encountered physiological conditions prior to purification and NMR analysis. To avoid copurification of chaperones with RNCs, all constructs were expressed in E. coli cells lacking ribosome-associated Trigger Factor and lysates were treated with ATP to release other chaperones such as GroEL or DnaK. The expression was performed in presence of 13C-glucose and/or 15N-NH4Cl, and stalled RNCs were isolated by centrifugation and affinity purification. RNCs were highly pure as judged by the Coomassie-stained gel showing exclusively the pattern of ribosomal proteins (Fig. S1A, lane 1). The signal of the nascent chains in the Western blot at 29 kDa demonstrated the homogeneity of the RNCs (Fig. S1B, lane 1). Moreover, the shift of this signal to a lower mass after Ulp1 treatment indicated quantitative cleavage of the N-terminal StrepII-Smt3-tag by the protease (Fig. S1B, lane 2). The processed RNCs carrying the SH3 domain were separated by centrifugation and subsequently used for recording the NMR spectra (Fig. S1 A and B lanes 3 and 4).

In Vivo Labeled Arrested RNCs Are Suitable for NMR Analysis.

To monitor the 3D structure of ribosome-associated full-length SH3 nascent chains (designated SH3-RNCs hereafter, Fig. 1B), [15N,1H]-CRINEPT-HMQC spectra of 15N-labeled SH3-RNCs as well as combined [15N,1H]-CRINEPT-HMQC/[13C,1H]-HMQC of 15N,13C-labeled SH3-RNCs were measured according to previous work (9, 15) (Fig. 2). Transverse and longitudinal relaxation-optimized NMR techniques (1619) enabled us to acquire NMR spectra with the spectral quality necessary for a detailed analysis within 20 h for the [15N,1H]-CRINEPT-HMQC and 3 h for the [13C,1H]-HMQC experiments. To prove that RNCs remained intact during data collection, we investigated their stability by recording a series of NMR spectra with 3 h measuring time. Additional cross peaks appeared in the NMR spectrum only after 48 h and in the spectral regions typical for C termini. These peaks are indicative of truncations of 15N-labeled nascent chains and hence deterioration of the RNC complex.

Fig. 2.
SH3 folds on ribosomes. [15N,1H]-CRINEPT-HMQC spectra of (A) SH3-F, (B) SH3-RNCs, (C) SH3 cleaved from the ribosome. [13C,1H]-HMQC spectra of (E) SH3-F, (F) SH3-RNCs, (G) SH3 cleaved from the ribosome. The peaks are labeled with the single amino acid ...

The recorded spectra of all ribosome-containing samples revealed additional cross peaks between 7.8 and 8.6 ppm in the [15N,1H]-correlation spectrum that neither corresponded to peaks of SH3 nor to the linker region (Fig. 2 AC and EG). Similarly, additional cross peaks were observed in the corresponding [13C,1H]-correlation spectrum. Because a cross peak at a typical location of C-terminal amides was also observed (strong cross peak at 8 ppm and 127 ppm; Fig. 2 B and C), we suspect that these cross peaks represent signals from 15N-labeled ribosomal proteins that had been synthesized and incorporated into ribosomes during the in vivo production of SH3-RNCs. This assumption is supported by the finding that such cross peaks were also measured for a crude ribosome sample generated under similar conditions as SH3-RNCs but without target nascent chains (Fig. S2). Attempts to adapt the labeling protocol according to a recently described method did not further reduce the background (10). However, the resulting [15N,1H]- and [13C,1H]-correlation spectra of the ribosome-tethered SH3-nascent chains are of similar high resolution as the recently published spectra of RNCs carrying the Ig2 domain, a tandem repeat segment of the gelation factor ABP120 from Dictyostelium discoideum (9).

To provide evidence that the labeled nascent polypeptides are stalled and associated with ribosomes, the sample was subjected to diffusion experiments using 1D 15N-filtered and 14N-filtered measurements. This allowed us to monitor the translational diffusion rates of 14N-containing ribosome as well as 15N-labeled nascent chain species. Both translational diffusion coefficients (Table S1) are similar to each other. These diffusion coefficients set a lower limit for the apparent molecular mass of both the 14N- and 15N-labeled species of ~1 MDa, thereby indicating that the nascent chains are attached to ribosomes. Some 15N-labeled background signals may in part contribute to the determined rates for SH3-RNCs. However, the translational diffusion coefficient of the 15N-labeled nascent chain was significantly higher upon TEV-protease-mediated release of SH3 from ribosomes, indicating that the labeled nascent chains were stably bound to ribosomes prior cleavage.

Comparing the NMR spectra of SH3-RNCs and SH3 cleaved from the ribosome (Fig. 2 B,C,F, and G) suggested interactions between the ribosome and the nascent chain. A detailed analysis of the spectrum shows that the average An external file that holds a picture, illustration, etc.
Object name is pnas.0914300107eq2.jpg line-widths of the resonances observed are 36 ± 2 Hz, which is much smaller than the expected value of > 1,000 Hz for the 70S ribosome particle, a line width well beyond detection (18). Release of ribosome-bound SH3 by TEV-cleavage led to the appearance of the C-terminal D63 of SH3 in the [15N,1H] correlation spectrum (Fig. 2C). The cleavage yielded improved 2D [15N,1H]-correlation and [13C,1H]-correlation spectra (Fig. 2C) with average An external file that holds a picture, illustration, etc.
Object name is pnas.0914300107eq3.jpg line widths of 28 ± 1 Hz. Although these line widths are smaller than those of SH3-RNCs, they are still considerably broader than the corresponding line widths of 15N-labeled SH3 free in solution in the absence of ribosomes (23 ± 1 Hz; Fig. S3). This difference is attributed to the high viscosity caused by the high concentration of ribosomes or RNCs (ranging from 100 to 200 mg/ml) in the samples. Furthermore, there are cross peaks that show similar intensities before and after the cleavage (for example, N48, L34 in Fig. 2B and C) indicating that the entire population of SH3-RNC was observed by NMR.

Ribosome-Associated Nascent SH3 Domain Adopts Native Conformation.

The assignment of SH3-RNCs (Fig. 2B) and SH3 domain released from ribosomes by proteolytic treatment (Fig. 2C) was based on the assignment of SH3 produced as a free domain called hereafter SH3-F (Figs. 1B and and22A). In addition, some linker residues of SH3-RNCs were identified (Fig. 2B) by measuring a released construct consisting of SH3 and a mutated SecM sequence carrying a point mutation that prevents stalling (SH3-SecM-F, Fig. 1B). This allowed for a residue-specific analysis of the 3D structures of SH3-RNCs with respect to both, the 3D structures of SH3-F and cleaved SH3. For the cleaved SH3 domain, the cross peaks of 55 15N-1H moieties and 93 13C-1H side chain moieties were unambiguously identified (Fig. 2 C and G). Although the spectral sensitivity for SH3-RNCs was lower compared to SH3-F and background signals from ribosomes were present in the spectra, a total of 22 15N-1H moieties and 23 13C-1H side chain moieties could be identified for specific residues of SH3-RNCs. These resonances were spatially well spread over the SH3 domain (Fig. 2 D and H), and the cross peaks of all these moieties were found at almost identical positions in the spectra with negligible average chemical shift deviations (ΔδHN of 0.040 ± 0.005 ppm and ΔδHC of 0.007 ± 0.002 ppm between SH3-RNC and cleaved SH3, and ΔδHN of 0.032 ± 0.005 ppm and ΔδHC of 0.011 ± 0.002 ppm between cleaved SH3 and SH3-F). Based on these results, we conclude that the nascent SH3 domain is competent to fold cotranslationally into a native structure, which is comparable to its conformation after release from the ribosome.

Ribosome-Associated SH3-m10 Shows a Random Coil-Like Conformation.

Next, we recorded the NMR spectra of SH3-m10-RNCs, which served as a reference for unfolded states of the SH3 domain (Fig. 3 and Fig. S4). As mentioned above, SH3-m10 harbors two point mutations that prevent native folding and instead generate a random coil-like state (13). The resonances obtained from [15N,1H]- and [13C,1H]-correlation spectra of free SH3-m10 (SH3-m10-F) revealed only a limited dispersion of chemical shifts reflecting the random coil structure of the domain (Fig. 3A and Fig. S4A). Due to the low degree of dispersion, only 54% of the side chains could be assigned, while the backbone assignment was nearly complete (97%). Furthermore, the 13C chemical shift deviations from corresponding “random coil” values were small, implying a lack of secondary structure elements (Fig. S4F). In particular, the huge overlap of resonances in the [13C,1H]-correlation spectrum can be attributed to the high flexibility of the side chains. Thus, there is little information content in the [13C,1H]-correlation spectrum, in striking contrast to the spectra of SH3 discussed above.

Fig. 3.
Nascent SH3-m10 has a random coil-like conformation. [15N,1H]-CRINEPT-HMQC spectra of (A) SH3-m10-F, (B) SH3-m10-RNCs, and (C) SH3-m10 cleaved from the ribosome. Peak labels are as in Fig. 2. (D) 1H and (E) 15N chemical shift deviations from the ...

Next, we measured the spectra of SH3-m10-RNCs following the same approach as for SH3-RNCs, including a sample in which the nascent chains had been released from the ribosomes by TEV-protease treatment (Fig. 3 B and C and Fig. S4 B and C). Using the sequential assignment of SH3-m10-F, a total of 55 15N-1H moieties and 43 13C-1H side chain moieties were unambiguously identified for released nascent SH3-m10. Similarly, 52 15N-1H moieties and 43 13C-1H side chain moieties were identified for SH3-m10-RNCs. The number of identified 15N-1H moieties for ribosome-associated SH3-m10 is considerably larger than for SH3-RNCs due to an increased signal to noise ratio attributed to the flexible nature of SH3-m10. Because of spectral overlap with background signals, only 33 15N-1H moieties and 15 13C-1H side chain moieties were used to analyze the SH3-m10-RNC structure. The cross peaks for these moieties in the NMR spectra of SH3-m10-RNCs, SH3-m10 cleaved from the ribosome and SH3-m10-F had almost negligible average chemical shift deviations (ΔδHN of 0.018 ± 0.001 ppm and ΔδHC of 0.006 ± 0.002 ppm between SH3-m10-RNCs and released SH3-m10; ΔδHN of 0.021 ± 0.002 ppm and ΔδHC of 0.147 ± 0.037 ppm between released SH3-m10 and SH3-m10-F).

Thus, these results confirm that ribosome-bound nascent SH3-m10 adopts a random-coil-like structure similar to SH3-m10 cleaved off of the ribosome and SH3-m10-F. Since the chemical shifts of random coil-like polypeptides are time-averaged over many fast interchanging conformations, the nearly negligible chemical shift deviations observed between SH3-m10-F and ribosome-tethered SH3-m10 nascent chains indicate that both polypeptides cover a similar conformational space with comparable residence times of their individual conformations.

Nascent SH3 Domain Remains Unfolded Until Its Entire Amino Acid Sequence Is Available.

An important question to be addressed is whether the SH3 domain folds cotranslationally in a two-state folding pathway as described for SH3 refolding in vitro or SH3 polypeptides adopt discrete intermediate conformations during elongation. In an attempt to mimic the ongoing translation of SH3, we designed two truncated versions of this domain named SH3-T1 and SH3-T2, lacking 22 and 5 C-terminal residues, respectively. SH3-T1 lacks the last three of the five β-strands, while in SH3-T2 only the last β-strand is absent. These truncated variants were produced as RNCs (SH3-T1-RNCs and SH3-T2-RNCs) and as free polypeptides designated SH3-T1-F and SH3-T2-F (Fig. 1B).

A total of 28 15N-1H and 44 13C-1H side chain moieties were unambiguously identified for SH3-T1-RNCs. A comparison of these with 31 15N-1H and 44 13C-1H side chain signals for SH3-T1-F in the presence of ribosomes allowed us to assess the folding status of the truncation variant (Fig. 4 AC and Fig. S5AC). Similarly, 25 15N-1H and 56 13C-1H side chain moieties could be identified for SH3-T2-RNCs as well as 41 15N-1H and 56 13C-1H side chain cross peaks for sample of SH3-T2-F mixed with ribosomes (Fig. 4 EG and Fig. S5 DF). The average chemical shift deviation between SH3-T1-RNCs and SH3-T1-F mixed with ribosomes was very small with ΔδHN of 0.016 ± 0.002 ppm and ΔδHC of 0.010 ± 0.002 ppm (Fig. 4 D and H). Similar results were obtained by comparing SH3-T2-RNCs and its soluble version SH3-T2-F (ΔδHN of 0.010 ± 0.002 ppm and ΔδHC of 0.035 ± 0.006 ppm). Importantly, the small spectral dispersion of SH3-T1 and SH3-T2 in the [15N,1H] and [13C,1H]-correlation spectra (Fig. 4 and Fig. S5), the random coil-like values of their 13C chemical shifts, and the close match between the shifts of SH3-T1, SH3-T2 and SH3-m10 (Fig. S5 K and L) clearly indicate that the ribosome-attached SH3-T1 and SH3-T2 as well as their soluble counterparts are in a random coil-like conformation.

Fig. 4.
Nascent SH3-T1 and SH3-T2 are unfolded. [15N,1H]-CRINEPT-HMQC spectra of (A) SH3-T1-F, (B) SH3-T1-RNCs, and (C) SH3-T1-F mixed with ribosomes. [15N,1H]-CRINEPT-HMQC spectra of (E) SH3-T2-F, (F) SH3-T2-RNCs, and (G) SH3-T2-F mixed with ribosomes. Peak ...

From these data we conclude that nascent SH3 does not cotranslationally adopt defined secondary or tertiary structure elements until the entire amino acid sequence of the domain is available. Thus, nascent SH3 folds domainwise and does not populate compact intermediate structures.

The Ribosome Confers No Major Constraints on Cotranslational SH3-Folding.

A recent NMR analysis suggests specific transient interactions between the ribosome and the folded ribosome-associated Ig2-domain (9). To obtain insights into the dynamic restraints on the nascent chain attached to the ribosome, we examined the line widths of the resolved resonances of the SH3-RNCs. We assumed that by binding to the ribosome the average rotational tumbling time of the SH3 would increase. We found that a number of cross peaks of SH3 nascent polypeptides bound to ribosomes exhibited differential line broadening when compared with SH3 released by TEV protease from ribosomes (Fig. S3) including eight 15N-1H cross peaks and nine 13C-1H cross peaks in the SH3-RNC spectrum that are broadened beyond detection. The residues of these resonances are spread throughout the entire SH3 fold and are solvent exposed (Fig. 2 D and H). In contrast, residues without significant line broadening were found in the interior of the protein. Importantly, considering that the line widths of the ribosome are > 1,000 Hz, the observed average line width increase of ~10 Hz by intermolecular interactions between the SH3 nascent chain and the ribosome indicates that less than 1% of SH3 interacts with the ribosome at any given time. Together with the finding that residues that interact with the ribosome are distributed over the entire SH3 surface, we conclude that these observed transient but very rare interactions are rather nonspecific and caused by random clashes due to tethering of SH3 to the ribosome.

We identified similar line broadening effects for the random coil-like nascent polypeptides investigated in our analysis (SH3-m10-RNCs, SH3-T1- and SH3-T2-RNCs, Figs. S4, S5. and S6). Line broadening was distributed along the entire amino acid sequence and detected for backbone and side chain moieties of both, hydrophobic and hydrophilic residues (Fig. S6). The line broadening effect was of similar amplitude as for SH3-RNCs, indicating that only a minor subpopulation of these nascent polypeptides interacts with the ribosomal exterior. Therefore, we conclude that there is no significant interaction between unfolded SH3 and the ribosomal surface that would contribute to the cotranslational folding of SH3.


Cotranslational folding is suspected to be critical for many newly synthesized polypeptides in order to prevent inter- or intramolecular misfolding and aggregation (20, 21). In this study, we set out to monitor the cotranslational folding pathway of the SH3 domain. Folding of denatured SH3 domains has been extensively studied by in vitro experiments showing that this domain folds quickly in a two-state transition (22). Recent studies revealed that the SH3 domain attains a protease-resistant conformation during its synthesis on ribosomes, suggesting that it folds cotranslationally into a compact structure (12). However, it remained unclear whether the folding pathway of nascent SH3 on ribosomes is similar to the refolding process of denatured SH3 in vitro. The incomplete sequence information of nascent SH3 during synthesis and the C-terminal fixation on ribosomes may influence the folding process, e.g., by populating intermediate conformations during the nascent state that are not formed during refolding of denatured full-length SH3 domain in solution.

The NMR studies revealed that the full-length SH3 domain tethered to the ribosome via SecM exhibits a compact, native-like β-sheet structure indicating that SH3 folds cotranslationally in vivo. In contrast, we found that shorter SH3 nascent chains remain unstructured during elongation and have a similar conformation to their free peptide counterparts in solution. Thus, the tethering of SH3 to the ribosome and its transient interactions with the ribosomal surface do not induce any thermodynamic stabilization of secondary and/or tertiary structures (Fig. 5). This implies that the cotranslational folding of SH3 resembles the two-state folding pathway observed during refolding of full-length SH3 in vitro. The folding of nascent SH3 on ribosomes is initially delayed until the entire sequence information is available for the productive cotranslational folding of the complete domain. Such a domainwise folding mode prevents off-pathway events that could become kinetic traps while allowing the sequential folding of the defined folding domains within the nascent polypeptide as it emerges from the ribosome. This mode of cotranslational folding might be particularly favorable for multidomain proteins. Interestingly, all RNCs tested herein had been expressed as polypeptides with two domains: an N-terminal Smt3-domain followed by the SH3 variants. Since only correctly folded Smt3 is recognized by the Ulp1 protease and all RNCs had been quantitatively cleaved during sample preparation, we conclude that also Smt3 folds cotranslationally and independently of the folding capacity of the downstream sequence. Whether this domainwise folding mode is specific for SH3 domains or holds true for other polypeptides with different domain folds remains to be tested.

Fig. 5.
Model of the domainwise cotranslational folding of nascent SH3. (A) Schematic model: The SH3 domain remains unfolded until the entire domain is synthesized and exposed outside of the ribosomal tunnel. (B) A scale model of cotranslationally folded native ...

How may the ribosome affect cotranslational protein folding? It is assumed that the ribosomal tunnel accommodates approximately 30 C-terminal residues of a growing nascent polypeptide. These amino acids cannot participate in structure formation of the polypeptide outside the tunnel and tethering of the nascent peptide may restrict its conformational freedom. Moreover, the ribosomal surface at the tunnel exit site might react with nascent chains exposed outside the ribosome. Indeed, a recent analysis reports transient interactions of a cotranslationally folded domain with the ribosomal surface (9). Although we observed transient interactions between the elongating SH3 polypeptides and the ribosome, these events appear to be very rare, involving only a minor population of the nascent chains at any given time. Based on our finding that neither the truncated SH3 constructs nor the unfolded SH3-m10 variant showed enhanced interaction with the ribosomal surface, we assume that these contacts can be attributed to nonspecific events that are exaggerated by the local tethering of the polypeptide to the translation machinery. Hence, the ribosomal surface itself confers no major constraints on cotranslational SH3 folding. Considering the small size of the SH3 domain nascent chain compared to the massive dimensions of the ribosome, it is even tempting to speculate that the ribosome may spatially shield emerging polypeptides from nonproductive interactions with cytosolic components (Fig. 5B). In bacteria, the ribosome-associated Trigger Factor complements this spatial confinement provided by the ribosome. This chaperone hunches over the polypeptide exit tunnel, thereby protecting nascent polypeptides against aggregation and degradation (12, 23). To get an even more comprehensive picture of nascent chains at the ribosomal exit site, the impact of Trigger Factor on the cotranslational folding of newly synthesized proteins has to be investigated at the atomic level.

Materials and Methods

Design, Labeling, and Purification of RNCs and Nonribosomal Constructs and Ribosome Preparation.

See SI Text.

NMR Spectroscopy.

All NMR spectra were recorded at 30 °C using a Bruker 700 MHz spectrometer equipped with four radio-frequency channels and a triple resonance cryoprobe with shielded z-gradient coil. The NMR samples contained either 15N-labeled or 13C,15N-labeled SH3 variants at a concentration of 0.01 mM in 10 mM 2H-BisTris(HCl)/95% H2O/5% D2O at pH 6.8. 15N,1H-CRINEPT-HMQC (18, 19) with and without 15N-decoupling were measured. In addition, also combined [15N,1H]-CRINEPT-HMQC/[13C,1H]-HMQC experiments (24) were measured. Typically, [15N,1H]-CRINEPT-HMQC experiments were recorded by using a 300 ms recycling delay with 1024 * 64 complex points (t1 max = 30 ms and t2 max = 100 ms) collected in the 1H and 15N dimensions, respectively, with a total of 6 * 128 transients per increment. Exponential and shifted sine window functions were used in the t1/F1 and t2/F2 dimensions, respectively. For the combined [15N,1H]-CRINEPT-HMQC/[13C,1H]-HMQC experiments similar values were used with a t1 max(15N) = 30 ms, t1 max(13C) = 7.5 ms and t2 max = 50 ms. In all the spectra quadrature detection in the indirect dimensions was achieved using States-TPPI (25). The residual water signal was suppressed by the WATERGATE sequence (26). For more details, see SI Text.

Supplementary Material

Supporting Information:


We thank Dr. J. Greenwald, Dr. K. Turgay, and members of the Deuerling and Riek labs for helpful discussions and comments on the manuscript. This work was supported by fellowships of the Konstanz Graduate School Chemical Biology and of the Zukunftskolleg to S. P. and grants of the Deutsche Forschungsgemeinschaft (DFG) to E. D. (DE-783/3-1).


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.0914300107/-/DCSupplemental.


1. Wegrzyn RD, Deuerling E. Molecular guardians for newborn proteins: ribosome-associated chaperones and their role in protein folding. Cell Mol Life Sci. 2005;62:2727–2738. [PubMed]
2. Ban N, et al. The complete atomic structure of the large ribosomal subunit at 2.4 Å resolution. Science. 2000;289:905–920. [PubMed]
3. Yusupov MM, et al. Crystal structure of the ribosome at 5.5 Å resolution. Science. 2001;292:883–896. [PubMed]
4. Ramakrishnan V. Ribosome structure and the mechanism of translation. Cell. 2002;108:557–572. [PubMed]
5. Seidelt B, et al. Structural insight into nascent polypeptide chain-mediated translational stalling. Science. 2009;326:1412–1415. [PMC free article] [PubMed]
6. Frydman J, Erdjument-Bromage H, Tempst P, Hartl FU. Co-translational domain folding as the structural basis for the rapid de novo folding of firefly luciferase. Nat Struct Biol. 1999;6:697–705. [PubMed]
7. Nicola AV, Chen W, Helenius A. Co-translational folding of an alphavirus capsid protein in the cytosol of living cells. Nat Cell Biol. 1999;1:341–345. [PubMed]
8. Berndt U, et al. A signal-anchor sequence stimulates signal recognition particle binding to ribosomes from inside the exit tunnel. Proc Natl Acad Sci USA. 2009;106:1398–1403. [PMC free article] [PubMed]
9. Hsu ST, et al. Structure and dynamics of a ribosome-bound nascent chain by NMR spectroscopy. Proc Natl Acad Sci USA. 2007;104:16516–16521. [PMC free article] [PubMed]
10. Rutkowska A, et al. Large-scale purification of ribosome-nascent chain complexes for biochemical and structural studies. FEBS Lett. 2009;583:2407–2413. [PubMed]
11. Nakatogawa H, Ito K. The ribosomal exit tunnel functions as a discriminating gate. Cell. 2002;108:629–636. [PubMed]
12. Hoffmann A, et al. Trigger Factor forms a protective shield for nascent polypeptides at the ribosome. J Biol Chem. 2006;281:6539–6545. [PubMed]
13. Blanco FJ, Angrand I, Serrano L. Exploring the conformational properties of the sequence space between two proteins with different folds: An experimental study. J Mol Biol. 1999;285:741–753. [PubMed]
14. Merz F, et al. Molecular mechanism and structure of Trigger Factor bound to the translating ribosome. EMBO J. 2008;27:1622–1632. [PMC free article] [PubMed]
15. Hsu ST, et al. Probing side-chain dynamics of a ribosome-bound nascent chain using methyl NMR spectroscopy. J Am Chem Soc. 2009;131:8366–8367. [PubMed]
16. Pervushin K, Riek R, Wider G, Wüthrich K. Attenuated T2 relaxation by mutual cancellation of dipole-dipole coupling and chemical shift anisotropy indicates an avenue to NMR structures of very large biological macromolecules in solution. Proc Natl Acad Sci USA. 1997;94:12366–12371. [PMC free article] [PubMed]
17. Pervushin KV, Wider G, Riek R, Wüthrich K. The 3D NOESY-[(1)H,(15)N,(1)H]-ZQ-TROSY NMR experiment with diagonal peak suppression. Proc Natl Acad Sci USA. 1999;96:9607–9612. [PMC free article] [PubMed]
18. Riek R, et al. Solution NMR techniques for large molecular and supramolecular structures. J Am Chem Soc. 2002;124:12144–12153. [PubMed]
19. Riek R, Wider G, Pervushin K, Wüthrich K. Polarization transfer by cross-correlated relaxation in solution NMR with very large molecules. Proc Natl Acad Sci USA. 1999;96:4918–4923. [PMC free article] [PubMed]
20. Deuerling E, Bukau B. Chaperone-assisted folding of newly synthesized proteins in the cytosol. Crit Rev Biochem Mol Biol. 2004;39:261–277. [PubMed]
21. Hartl FU, Hayer-Hartl M. Molecular chaperones in the cytosol: From nascent chain to folded protein. Science. 2002;295:1852–1858. [PubMed]
22. Cobos ES, et al. A thermodynamic and kinetic analysis of the folding pathway of an SH3 domain entropically stabilised by a redesigned hydrophobic core. J Mol Biol. 2003;328:221–233. [PubMed]
23. Ferbitz L, et al. Trigger Factor in complex with the ribosome forms a molecular cradle for nascent proteins. Nature. 2004;431:590–596. [PubMed]
24. Mueller L. Sensitivity enhanced detection of weak nuclei using heteronuclear multiple quantum coherence. J Am Chem Soc. 1979;101:4481–4484.
25. Marion D, Ikura M, Tschudin R, Bax A. Rapid recording of 2D NMR spectra without phase cycling. Application to the study of hydrogen exchange in proteins. J Magn Reson. 1989;85:393–399.
26. Piotto M, Saudek V, Sklenář V. Gradient-tailored excitation for single-quantum NMR spectroscopy of aqueous solutions. J Biomol NMR. 1992;2:661–665. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...