Co-translational mechanisms of quality control of newly synthesized polypeptides

Introduction

Protein synthesis is one of the most energy-consuming processes in the cell¹ and is a major step in the regulation of gene expression.² Notwithstanding the relatively high fidelity of mRNA translation (approximately 2–3 errors for every 10,000 amino acids added to a NSP in mammalian cells),³ rates of incorrect amino acid incorporation arise in an estimated 20% of proteins due to an accumulation of errors in protein synthesis and/or upstream gene expression regulatory mechanisms (e.g., transcription).^4,5 In addition, aberrant proteins can be generated by anomalous post-translational modifications and dysfunction of trans-acting factors that influence the folding of NSPs.⁵ Eukaryotes have evolved a variety of mechanisms to ensure high-quality surveillance of their proteomes. Accordingly, a substantial proportion of NSPs are thought to be rapidly degraded during or shortly after their synthesis by the ubiquitin-proteasome system.^6,7 The exact percentage of NSPs that are degraded remains the subject of debate. Pioneering studies contend that approximately 30–50% of NSPs are co-translationally degraded.^8,9 However, cells in these studies were deprived of amino acids prior to metabolic labeling in order to minimize competition between the radiolabeled and endogenous unlabeled amino acids. Subsequently, it was demonstrated that amino acid deprivation induces a feedback loop mechanism, whereby proteasomes are activated to degrade mature, non-aberrant proteins to replenish the endogenous amino acid pool.¹⁰ This observation strongly suggests that the proportion of degraded NSPs was overestimated due to methodological shortcomings.¹⁰ More recently, the extent of NSP degradation was assessed using an alternative method that centers on the level of ubiquitination in lieu of metabolic labeling. Here, nascent polypeptide chains were labeled with a biotin-conjugated fluorescent version of puromycin.¹¹ Using this approach, the authors have found that approximately 15% and 8% of NSPs are co-translationally ubiquitinated in mammalian cells and in yeast, respectively.¹¹ Taken together, these findings indicate that a sizable portion of NSPs are co-translationally ubiquitinated and degraded by the proteasome.

In addition to their well-established role in mRNA translation, ribosomes play a central role in the regulation of folding and degradation of NSPs. Herein, we will review recent findings that illuminate the molecular mechanisms underlying the role of the ribosome and other components of the translational machinery in the quality control of NSPs and discuss the functional consequences of these processes.

The role of ribosomes in the folding of NSPs

As they emerge from the ribosomes, nascent polypeptide chains fold into complex three-dimensional structures that characterize the functional protein.¹² The dimensions of the ribosome exit tunnel are limiting (80–100Å long and 10–20Å wide in yeast),^13,14 only allowing folding of the elongating polypeptide chain up to the secondary structural elements.¹⁵ Consequently, the final stages of NSP folding must be accomplished in the crowded cytoplasmic environment. Nascent polypeptides that emerge from the ribosome contain surface-exposed hydrophobic domains that can engage in nonspecific interactions with an array of cytosolic proteins, as well as with the other nascent polypeptides.^16,17 Ribosome-associated chaperones protect these partially folded polypeptides from these nonspecific interactions, thereby increasing the likelihood of accurate folding.¹⁸

The growing nascent chain first encounters the nascent polypeptide-associated complex (NAC).¹⁹ The NAC is a heterodimeric complex that associates with ribosomes in close proximity to the ribosome exit tunnel (Fig. 1A).¹⁹ It shields approximately 30 N-terminal amino acids of the nascent polypeptide.¹⁹ As the nascent polypeptide chain lengthens, the NAC is replaced by other ribosome-associated factors, such as the signal recognition particle (SRP) and the elongation factor eEF1A, both of which have a higher affinity for the hydrophobic regions of the nascent polypeptide chain than the NAC (Fig. 1A).²⁰ Subsequently, additional chaperones are recruited to the exposed and partially folded hydrophobic domains of NSPs. Heat shock proteins Hsp70, Hsp90 and the eukaryotic TCP-1 ring complex (TRiC) were identified as the predominant classes of these chaperones (Fig. 1A).^21-25 In eukaryotic cells, Hsp70 chaperones associate with ribosomes, bind elongating polypeptide chains and facilitate their folding, which promotes the association of NSPs with downstream chaperones such as Hsp90²⁶ or TRiC.^21,27 Hsp70 chaperones preferentially promote the folding of long nascent chains that are characterized by abundant hydrophobic regions.²³ Although some NSPs that had been partially folded by Hsp70 are transferred to Hsp90,²⁶ the role of Hsp90 chaperones in co-translational folding remains largely elusive. Hsp90 chaperones appear to regulate the folding of a restricted subset of proteins such as steroid receptors and Src-related kinases.²⁸ TRiC is a large multimeric complex that bears a resemblance to the bacterial GroEL double ring.²⁹ It was originally identified as a chaperone complex that binds to polar regions of newly synthesized cytoskeletal proteins and facilitates their optimal folding.^30,31 More recently, the TRiC interactome has been further explored.³² Like its prokaryotic homolog GroEL, the TRiC complex appears to associate with a plethora of structurally and functionally distinct NSPs, which indicates that it plays a general role in the folding of a broad spectrum of NSPs.^32,33

In yeast, NAC cooperates with Hsp70 chaperone Stress 70B (SSB)–Ribosome Associated Complex (RAC) and the Hsp110 chaperone Sse1 in the folding of NSPs.³⁴ RAC is a heterodimeric complex formed by members of the conserved Hsp70 (Ssz) and Hsp40 chaperone families (Zuotin).³⁵ Zuotin interacts with rpL31,²⁵ which surrounds the peptide exit tunnel.³⁶ In yeast, deletion of Sse1 and SSB results in protein aggregation, which is further exacerbated by the loss of NAC.³⁴ Intriguingly, depletion of Sse1 and SSB also causes reduction of small and large ribosomal subunits with concomitant enrichment of ribosomal proteins and rRNA in the pool of aggregated NSPs. This suggests the presence of a regulatory feedback loop whereby an increase in NSP aggregation in cells with abrogated chaperone activity downregulates protein synthesis, thereby preventing further accumulation of misfolded NSPs.³⁴ These findings show that RAC and associated chaperones play an important role in the optimal folding of NSPs and moreover, that they serve as a central node of protein quality control, coordinating both protein folding and rates of mRNA translation (see below).

Ribosomes Link Translational Control and NSP Folding

Folding of nascent polypeptides may be coordinated with translation rates, which in turn are determined by both protein length and mRNA sequence.^37,38 Mild inhibition of protein synthesis has been shown to facilitate folding of NSPs.³⁹ Interestingly, in yeast, proteins that are synthesized rapidly are more prone to co-translational ubiquitination, an effect that appears to be mediated by a complex network of E3 ubiquitin ligases, including ribosome-bound Rkr1 and Hel2.⁴⁰ In general, the translation rate and folding efficiency is determined, at least in part, by features of the mRNA that are being translated. For example, yeast TRP3 mRNA contains a cluster of consecutive rare codons that correspond to a highly-structured region of PYK protein that separates its two enzymatic domains. ⁴¹In light of these and similar findings, it has been proposed that mRNA regions encompassing elongation pausing sites define the domain boundaries and facilitate the structural organization of NSPs.^42-45 Potential effects of codon bias on protein folding also highlight hitherto unappreciated effects that synonymous single-nucleotide polymorphisms (SNPs) may exert on protein function. Namely, SNPs are commonly not expected to affect synthesis, folding and function of polypeptides as the genes harboring them encode the same amino acids as their wild-type counterparts. However, it has been demonstrated that SNPs in the multidrug resistance (MDR1) gene that encodes transmembrane ABC-transporter MDR1/P-glycoprotein affect its folding and transporter activity.^46,47 In addition, an elongation pause site in signal recognition particle receptor (SRα) mRNA is required for co-translational ER membrane anchoring of SRα receptor, inasmuch as synonymous mutations that alleviate elongation pausing impede ER localization of this receptor.⁴⁸ Therefore, optimal protein function requires well-orchestrated regulation of translation and protein folding.

Ribosome-Dependent Regulation of Ubiquitination and Proteasomal Degradation of NSPs

Misfolded NSPs must be rapidly targeted for degradation to avoid the accumulation of protein aggregates that have deleterious repercussions for cells. Accordingly, several key components of the ribosome-associated NSP quality surveillance system have been recently described including Listerin (Ltn1 in yeast), translation elongation factor 1A (EF1A), and 97-kDa valosin-containing protein (p97/VCP; Cdc48 in yeast) (Fig. 1A and B).

The Cross-Talk between Ribosomes and Chaperones Determines Translation Rates under Stress

Cells respond to various types of stress, such as UV-irradiation, nutrient deprivation and ER-stress, by reprogramming gene expression and shutting down global mRNA translation.⁸⁰ A stress-induced decrease in the global protein synthesis rate is mediated by the members of the eukaryotic translation initiation factor 2 (eIF2) kinase family (PERK, PKR, GCN2 and HRI).⁸¹ eIF2 forms a ternary complex with the initiator tRNA (Met-tRNAi) and GTP that delivers Met-tRNAi to the 40S ribosomal subunit to form the 43S pre-initiation complex.⁸² At the end of initiation, eIF2-bound GTP is hydrolyzed and the binary eIF2/GDP complex is released from the ribosome.⁸³ eIF2B is a guanine nucleotide exchange factor (GEF) that exchanges GDP bound to eIF2 for GTP allowing the formation of the ternary complex and another round of initiation.⁸³ eIF2 kinases inactivate eIF2 via phosphorylation of its regulatory eIF2α subunit, which stabilizes eIF2/GDP/eIF2B complexes and inhibits GEF activity of eIF2B.^84-86 Whereas phosphorylation of eIF2α inhibits global translation, it promotes the translation of a specific subset of mRNAs harboring short, inhibitory upstream ORFs (uORFs) that encode master transcriptional regulators of stress response such as CHOP and ATF4 (Fig. 2A).⁸⁷ In non-stressed cells, uORFs impede the progression of scanning ribosomes toward the downstream initiation codon.^88,89 In contrast, stress-induced eIF2α phosphorylation attenuates re-initiation on uORFs, thereby favoring the translation of downstream ORFs.⁹⁰

In addition to the well-established role of eIF2α phosphorylation in stress-induced downregulation of mRNA translation, several recent studies implicate chaperones in this process. Proteotoxic stress is a condition in which protein homeostasis (protein synthesis, folding, and clearance) is altered by various stressors including heat-shock or oxidation.⁹¹ Cells respond to proteotoxic stress by decreasing mRNA translation rates and increasing the chaperone capacity.⁹² Recent findings show that in addition to its role in the folding of NSPs, Hsp70 bolsters translation elongation rates by facilitating extrusion of nascent chains from the ribosome exit tunnel.^21,93 In cells exposed to heat stress or toxic amino acid analogs, Hsp70 chaperones dissociate from ribosomes, presumably due to excess misfolded proteins out-competing nascent polypeptides for Hsp70. The overall reduction in ribosome-associated Hsp70 is paralleled by a genome-wide elongation pausing ~100 nucleotides downstream of the initiation codon.^93,94 Under these conditions, stalled ribosomes shield the first ~30–40 amino acids of nascent polypeptide chains (Fig. 1A). Consistently, depletion or pharmacological inhibition of Hsp70 recapitulates the elongation pausing phenotype even in the absence of proteotoxic stress.⁹³ Therefore, elongation pausing of ribosomes is considered to be a strategy to prevent potentially harmful unspecific interactions of nascent polypeptides, and to allow time for more Hsp70 chaperones to become available, as well as to block other ribosomes from engaging the mRNA under conditions where chaperone capacity is diminished.^93,94

Although the mechanism underlying the pausing of ribosomes during elongation remains poorly understood, emerging data indicate that chaperones and the ribosome exit tunnel collaboratively orchestrate protein folding and ribosome scanning (Fig. 1B).⁹⁵ Ribosomal proteins of the large subunit rpL4 and rpL22 appear to form what resembles a gated opening in the ribosome exit tunnel³⁶ and have a role in ribosome stalling.¹⁵ Hsp70 chaperones associate with rpL22 and rpL4 and this interaction may cooperate to resize the ribosome exit tunnel.^36,93 These studies provide the first look into the mechanisms that chaperones may employ to coordinate protein folding and translation elongation rates.

NSP Quality Control during Stress

Although ribosome-mediated quality control of NSPs represents an important mechanism of proteome surveillance, mechanisms that regulate this process in response to stress remain largely obscure. Below, we discuss recent findings that provide the first evidence implicating the mammalian target of rapamycin complex 1 (mTORC1) and c-jun N-terminal kinase (JNK) in the regulation of NSP folding and degradation under stress.

Future Challenges and Concluding Remarks

Emerging data indicate that, in addition to their well-established role in translation, ribosomes and other components of the translational machinery play a major role in NSP surveillance. However, cellular networks that link protein synthesis, chaperones, and proteasome machineries are still largely unknown. Alterations in chaperone capacity modulate protein synthesis rates,^94,93 whereas inhibition of translation appears to downregulate chaperone production by disrupting the transcriptional activity of heat shock factor protein 1 (HSF1).¹¹³ In turn, HSF1 has been shown to regulate mTORC1 activity during stress and to stimulate rDNA transcription in cancer cells, thus suggesting that it may bolster translation.^99,114 In addition to HSF1, SKN-1 (C. elegans ortholog of mammalian nuclear respiratory factors Nrf1, 2 and 3), a master transcriptional regulator of oxidative stress, has been shown to coordinate rates of protein synthesis and degradation by modulating the activity of the ubiquitin-proteasome system.¹¹⁵ Inhibition of the proteasome leads to the induction of SKN-1, whereas suppression of translation leads to its inactivation.¹¹⁵ Importantly, SKN-1 promotes resistance to oxidative stress and extends the lifespan of worms.¹¹⁶ These findings suggest that HSF1 and SKN-1 may serve as central nodes of cellular stress network that sense alterations in translation rates and proteotoxic stress to coordinate protein synthesis, folding and decay to maintain organismal and cellular homeostasis.

In addition to knowledge gaps related to the molecular underpinnings of cellular networks that coordinate mRNA translation, NSP quality control and degradation, it is still not clear whether NSP surveillance equally affects all cellular proteins, or whether it targets specific subsets of NSPs. The potential importance of selective mechanisms of NSP surveillance is substantiated by defective ribosomal products (DRiPs) that are generated by rapid degradation of a specific subset of NSPs shortly after completion of their synthesis and serve as peptide ligands for MHC class I molecules.^117,118 These findings suggest that targeted degradation of NSPs may play a major role in essential cellular processes such as immune response and immunosurveillance of tumors.¹¹⁹ The plausibility of the selective degradation of NSPs also puts forward a model, whereby decay of a subset of NSPs would produce changes in the proteome that are required for the optimal stress response (e.g., by degrading nascent polypeptides that are not required during stress). In this way, degradation of NSPs may serve not only as a quality control mechanism, but also as the crucial step in the regulation of gene expression.

In this review, we summarized emerging mechanisms of co-translational surveillance of NSPs and the networks that coordinate translation rates, folding and degradation of NSPs. The results of these studies strongly suggest that the co-translational surveillance of NSPs is achieved via the orchestrated action of the translational machinery, chaperones and the ubiquitin-proteasome system.