An Emerging Role for the Endoplasmic Reticulum in Stress Granule Biogenesis

Abstract

Stress granules (SGs), structurally dynamic, optically resolvable, macromolecular assemblies of mRNAs, RNA binding proteins (RBPs), translation factors, ribosomal subunits, as well as other interacting proteins, assemble in response to cell stress conditions that elicit phosphorylation of eukaryotic initiation factor 2α (eIF2α) and consequently, the inactivation of translation initiation. SG biology is conserved throughout eukaryotes and has recently been linked to the pathological sequelae of neurodegenerative disorders, cancer biology, and viral infection. Substantial insights into mechanisms of SG biogenesis, and more broadly the phenomenon of biological liquid-liquid phase separation (LLPS), have been aided by detailed proteomic and transcriptomic studies as well as in vitro reconstitution approaches. A particularly interesting and largely unexplored element of SG biology is the cell biological context of SG biogenesis, including its subcellular organization and more recently, evidence that the endoplasmic reticulum (ER) membrane may serve important functions in RNA granule biology generally and SG biogenesis specifically. A central role for the ER in SG biogenesis is discussed and a hypothesis linking SG formation on the ER to the trafficking, localization and de novo translation of newly exported mRNAs is presented.

1. Introduction

1.1. Stress Granules and the Integrated Stress Response

Stress granules (SGs) are a prominently studied member of a family of ribonucleoprotein (RNP) granules evident at many phases of mRNA biology from transcription, to splicing, localization and translation [1–6]. As higher order, multicomponent assemblies of proteins and mRNAs, SGs are a form of biomolecular condensate that has attracted considerable research interest and provides a leading example of liquid-liquid phase separation as a subcellular organization and regulatory principle [2, 7–10]. As a field, the study of biomolecular condensates has leveraged principles of polymer physics and in particular the physicochemical basis of liquid-liquid phase separation to characterize the complex, dynamic behavior of RNP granules in non-equilibrium biological systems [9, 11–15]. Although compositionally distinct, SGs share with many RNP granules three common properties; they are non-membranous assemblies, contain RNA, largely in the form of untranslated mRNAs, and are enriched in RNA binding proteins (RBPs), many with intrinsically disordered and/or prion-like domains that support the multimerization behavior critical to macromolecular phase separation [2, 8, 16–18]. SGs can be distinguished from many RNP condensates in that, eponymously, their formation is elicited by diverse cell stress conditions and most prominently, but not exclusively, cell stress conditions that promote elongation factor 2α (eIF2α) phosphorylation and the consequent inhibition of the initiation stage of protein synthesis [4, 19–22]. Notably, SGs biogenesis can also be elicited via inactivation of eIF4A, eIF4B, eIF4G, eIF4H, or PAPB as well as small molecule eIF4A inhibitor pateamine A, further validating the mechanistic coupling of protein synthesis initiation with SG biogenesis [23–25].

eIF2α phosphorylation is mediated by four kinases, GCN2, PKR, HRI, and PERK, each of which is responsive to distinct cellular stresses [26, 27]. GCN2 (general control non-derepressible 2) binds deacylated tRNAs and thus monitors steady state amino acid/charged tRNA levels [28, 29]. PKR (protein kinase R) is activated by binding to double-stranded RNA and serves a primary role in the cellular response to virus infection; PKR can also be activated by several cellular stressors, including serum starvation, and peroxide and arsenite treatment, via the protein activator PACT [28, 30–32]. HRI (heme-regulated kinase) is responsive to iron-heme levels and undergoes activation in response to sodium arsenite, an environmental toxin commonly used in studies of SG biology [33, 34]. PERK (PKR- like ER kinase) is a resident ER transmembrane kinase with an ER lumenal unfolded protein sensing domain and a cytosolic eIF2α kinase domain [35, 36]. PERK activation occurs in response to the accumulation of unfolded proteins in the ER and can be elicited by a number of pharmacological disruptors of proteostasis including tunicamycin, an inhibitor of N-linked glycan biosynthesis which promotes nascent N-linked glycoprotein misfolding [37, 38], reducing agents such as dithiothreitol (DTT) or β-mercaptoethanol (BME), which prevent disulfide bond formation in the ER lumen and thereby disrupt protein folding [39, 40], and thapsigargin, which disrupts ER proteostasis via inhibition of the sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase SERCA [41, 42]. The mechanistic coupling of eIF2α phosphorylation to SG formation thus places SG biogenesis in the compendium of cellular responses to activation of the integrated stress response (ISR) (Fig.1). In further support of protein synthesis initiation inhibition as the critical driver of SG biogenesis, pharmacological inhibition of eukaryotic initiation factor 4A (eIF4A), an RNA helicase, by the compounds pateamine A or hippuristanol also stimulate SG formation [24]. Additionally, links between protein synthesis initiation inhibition and SG biogenesis are supported by the discovery that ISRIB (integrated stress response inhibitor), a small molecule inhibitor of the ISR that selectively binds and stabilizes active higher order oligomeric states of the dedicated eIF2 guanine nucleotide exchange factor eIF2B, blocks de novo SG formation and promotes the rapid dissolution of SGs when supplied to stressed cells [43]. Notably, ISRIB blocked sodium arsenite-induced SG formation and stimulated SG resolution when provided to cells pretreated with sodium arsenite or the pharmacological inducers of the unfolded protein response (UPR) tunicamycin and thapsigargin, which as noted above promote PERK activation [43]. The latter is of particular interest as the UPR selectively reports on the protein folding environment of the ER lumen.

Figure 1. Cell stress elicits translational suppression through activation of eIF2α kinase activity.

Cell stress, sensed as disruptions in ER luminal proteostasis (via PERK), the accumulation of deacylated tRNAs (via GCN2), the accumulation of reactive oxygen species (ROS) or nitric oxide (NO) (via HRI), or pathogen- or host-derived double stranded RNA (via PKR) promotes kinase dimerization and subsequent activation. Kinase activation results in phosphorylation of eIF2α disabling eIF2 interactions with its cognate GEF eIF4B, and the suppression of translation initiation. Consequently, polyribosomes undergo run-off translation and protein synthesis in both the cytosol and endoplasmic reticulum compartments of the cell is inhibited.

A principal function of the UPR is to transiently suppress the translation of ER-targeted mRNAs, thereby reducing the flux of nascent polypeptides into the ER lumen. Intriguingly, RNA-seq analyses of purified SGs revealed that ER-targeted mRNAs are under-represented in the SG mRNA transcriptome, suggesting that localization to the ER compromises mRNA recruitment to SGs even when translation initiation is suppressed [44]. This finding may have a relatively simple technical explanation, where current methods for SG isolation are selective for soluble populations of SGs, as discussed by these authors [44]. The question of ER-targeted mRNA recruitment into SGs does however appear to be complex. For example, MDR1 transcripts, encoding the polytopic integral membrane protein multidrug resistance P-glycoprotein, were demonstrated to be refractory to SG recruitment [45]. Yet, disruption of the translation and ER-localization of MDR1 mRNAs, via insertion of a stable stem-loop in the 5’ UTR of a truncated MDR1 reporter, enabled SG recruitment of the (cytosolic) MDR1 reporter during arsenite stress [45]. As will be later discussed, recent evidence supports a role for the ER, perhaps fundamental, in SG biogenesis. Of particular importance, there is evidence, albeit limited, that mRNA recruitment into SGs may have a gene specificity element where ER-localization per se does not preclude mRNA recruitment into SGs but rather ER-localized mRNAs display variations in their sensitivity to SG recruitment [46].

2. SG Proteomics Identify Constitutive Low Assembly State Granules.

Efforts towards a comprehensive understanding of SG protein composition and function via biochemical purification, in situ proximity proteomics, and domain/motif analysis have revealed that SGs are enriched in proteins containing intrinsically-disordered domains (IDRs) and other relevant structural features such as RRM and PARP domains, prion-like domains, and RGG/RG motifs, that support RNA binding and protein-protein interactions conducive to phase separation behavior [8, 16, 17, 20, 47, 48]. An intriguing and significant finding from the in situ proximity proteomics experimental approaches was the identification of SG protein-protein interaction networks in unstressed cells [17, 49]. That such networks are present under homeostatic conditions favors the interpretation that these interaction networks are integral to the biology of mRNA export, quality control, localization, and translation. In this view, SGs are higher order assemblies of protein-protein interaction-RNA regulatory networks operating at homeostatic steady-state, rather than a unique, de novo macromolecular assembly process occurring in response to the imposition of cell stress and in particular, activation of the ISR. In this view, cell stress-evoked inhibition of translation initiation elicits a kinetic blockade of, for example, mRNA quality control pathways and as a result RNP pathway intermediates elaborate into higher order structures of dimensions detectable by light microscopy [50–53]. Building on this view, it is of interest that many SG-resident RBPs are shuttling proteins that accompany newly transcribed mRNAs as they exit nuclear pores and following translation-driven RBP remodeling are shuttled back to the nucleus. Indeed, structural features that are enriched in SG RBPs, including RG/RGG and RSY motifs, function as non-classical nuclear localization signals and are recognized by the nuclear import factors transportin-1 and transportin-3 (karyopherin β2, importin α/β) [54–57]. Furthermore, these protein-protein interactions are deterministic in that when these motifs are complexed with the requisite nuclear import factors the proteins undergo nuclear import whereas in the absence of such interactions, protein-protein interactions conducive to phase separation are enabled [54, 55, 57]. Also consistent with this view is the finding that cell stress evokes the accumulation of nucleocytoplasmic transport factors in SGs and disruptions in nucleocytoplasmic protein transport [58, 59]. These findings highlight a functional link between nucleocytoplasmic transport and SG biogenesis, as schematically elaborated in Fig. 2.

Figure 2. Schematic illustration of a model linking stress granule biogenesis to the translational remodeling of newly exported mRNAs.

In this model, newly exported mRNAs are depicted as the primary mRNA substrates for SG biogenesis. Under conditions of cellular homeostasis, newly exported mRNAs are engaged by membrane-bound (either ER or outer nuclear envelope, which are physically continuous) translation machinery where they undergo the initial/pioneering rounds of translation. These early translation events are accompanied by extensive remodeling of RNA binding protein (RBP) composition, where predominately nuclear RBPs, many with the capacity for multimerization, are displaced to undergo shuttling back to the nucleus. This scanning process is thought to be coupled to nonsense-codon scanning, miRNA silencing, and other translational control events that are initiated on newly exported mRNAs. As depicted, mRNAs that complete these quality control remodeling and scanning processed are released as “mature” mRNAs that then undergo productive translation. Because of the loss of a nuclear RBP signature/gain of a cytoplasmic RBP signature, these “mature” mRNAs are largely refractory to stress granule recruitment. During cell stress conditions where translation initiation is suppressed, newly exported mRNAs retain a nuclear RBP signature which confers a high susceptibility to SG recruitment. The continued suppression of translation initiation and growing supply of newly exported, non-translated mRNAs supports higher order, liquid-liquid phase separation-driven formation of large (optically detectable) granules. In this model, SG formation is depicted as occurring on the cytosolic face of the ER membrane, in complex with resident ER integral membrane proteins which through interactions with mRNAs, ribosomal subunits, translation machinery, and other components of the SG proteome support ER-associated SG biogenesis.

3. Regulatory Intersections Between Membranous and Non-Membranous Organelles.

The growing interest in macromolecular phase separation as a biological organizing principle has fostered highly interdisciplinary research into SG biology and yielded experimental and theoretical frameworks for understanding SG biogenesis [9, 12, 13, 15, 50]. A largely unexplored research area in SG biology concerns the intersection of SG biogenesis and the complex ultrastructure of eukaryotic, and more specifically metazoan cells. Do SGs form de novo via random stochastic interactions? Alternatively, do SGs elaborate from existing RNP-protein complexes? Do such complexes diffuse freely in the cell or are they anchored to specific cellular subdomains? An interesting and possibly related link extends from proximity proteomic studies of SG composition noted above, where the protein interactome networks of core SG components such as G3BP1 were both present at homeostasis and only modestly remodeled in different stress scenarios [52, 60]. As raised by these authors and noted above, SGs are higher order forms of ribonucleoprotein assemblies present at steady state, though at physical scales below the resolution limits of current optical imaging methods [17, 52].

A particularly intriguing cell biological question concerns the subcellular site of SG biogenesis, in particular the question of whether SGs assembly is 3D/solution-based (i.e., cytosolic) and/or if SGs biogenesis can occur on membranes such as the endoplasmic reticulum. Regarding the latter, it’s of interest to determine if the kinetic advantages of dimensionality reduction provided by membrane-linked assembly might contribute to SG biology. To the first point, a role for the ER in the regulation of membraneless organelle dynamics was first reported by the Voeltz and Parker labs [61]. Using a dimerization-dependent fluorescent protein (ddFP) domain interaction experimental approach with a red fluorescence-capable (RA) domain-ER translocon component Sec61β chimera and a core (GB) domain chimera with the processing body (PB) component Dcp1b, these investigators noted that a substantial fraction of PBs were ER-localized for time periods of > 2 min and in analyses of independent tracking dynamics, the ER and PBs were equivalently co-localized in live cell imaging studies [61]. Though the majority of the data reported by these authors concerns PB-ER colocalization and ER-linked regulation of PB fission and fusion, these authors also reported a role for the ER in SG fission occurring during ISRIB-induced SG resolution [61]. These findings implicate the ER in nonmembranous organelle fusion and fission dynamics, where ER-organelle/nonmembranous organelle contact sites regulate granule dynamics, either in the regulation of granule growth (fusion) and/or dissolution (fission). At present the molecular identities of the membrane components, presumably ER-resident membrane proteins, that mediate interactions with PB components are unknown. Experimental evidence supporting a role(s) of the ER in the regulation of nonmembranous organelle assembly was also recently reported in studies of the fungal RNA binding protein Whi3. Whi3, which performs key functions in the yeast pheromone response, has been shown to assemble into granule-like assemblies that interact with/co-localize with the ER [62]. Consistent with this subcellular localization, studies of the Whi3 RNA interactome demonstrated a significant enrichment for ER-localized mRNAs, in particular those encoding proteins localized to the ER, plasma membrane, and cell wall [63]. An intriguing role for the ER in the regulation of Whi3 granule assembly dynamics was recently reported by the Gladfelter lab [64], who identified a regulatory role for Whi3-ER membrane interactions in the control of Whi3-RNA condensate size [64]. Using in vivo imaging approaches in the multinuclear fungi Ashbya gossypii as well as in vitro reconstitution studies on 2D planar supported lipid bilayers, these authors identified novel roles for Whi3-membrane association in limiting condensate assembly size (e.g., coarsening) and attributed this phenomenon to restricted diffusional mobility, postulating a role for the Whi3-ER membrane association in the kinetic control of condensate size [64]. In this model, the Whi3-ER membrane association serves a regulatory function where membrane localization favors condensate formation but imposes diffusional constraints on condensate growth. This phenomenon may also be part of a complex process where Whi3 partitioning between the cytosol and ER compartments is itself regulated, perhaps via association with ER-targeted mRNAs as occurs in yeast during the control of cell mating fate decisions [62, 65].

The studies noted above provide multiple lines of experimental evidence supporting a regulatory role for the ER in the control of nonmembranous organelle dynamics and illustrate scenarios where the localization of granule assembly to the ER can be mechanistically linked to cell state/ fate outcomes and where granule size can be regulated by virtue of its membrane localization. How contact site-based interactions of the ER with PBs serves to regulate PB fission and as importantly, how ER-regulated PB fission impacts PB function remain to be determined. PBs and SGs do share numerous (> 100) proteins and although there is substantial evidence for unique compositional and functional elements to PB and SG biology, the overlapping composition and stress-linked assembly dynamics are sufficient to consider a role for the ER in SG biology.

Further motivating investigation into potential links between the ER and SG biology are the findings noted above that ER-targeted mRNAs are under-represented in the SG transcriptome and that ER-localization is associated with an mRNA being refractory to recruitment into SGs [44, 45]. Although the available data is largely consistent with this conclusion, there remains a conundrum in the view that localization of mRNAs functionally segregates them from SG recruitment. In part, numerous studies have demonstrated that the mRNA transcriptome is broadly represented on the ER and on average approximately 1/3^rd of the population of a given cytosolic protein-encoding mRNA undergoes translation on ER-bound ribosomes [66–69]. Thus, essentially all mRNAs undergo at least fractional translation on the ER regardless of the subcellular destination of their translation product. If ER localization precludes or suppresses mRNA recruitment to SGs, it follows that only those mRNAs engaged on cytoplasmic, free ribosomes are accessible for SG recruitment following inhibition of translation initiation. In this view, and although the ER serves a regulatory function in PB and potentially SG fusion/fission dynamics, that function is distinct from the processes that determine mRNAs susceptibility to SG assembly.

Contrasting this view, a recent study from the author’s lab examined functional links between the ER and SG biogenesis and reported that SGs form in close apposition to and/or in direct association with the ER, and in response to both pharmacological activation of the unfolded protein response or the addition of sodium arsenite [46]. Although the optical imaging approaches used in that study were of insufficient resolution to reveal if SG biogenesis occurs on the ER, in situ cell fractionation experiments provided additional evidence in support of the ER as the subcellular site of SG assembly. In these experiments, SGs were imaged via both single molecular fluorescence in situ (smFISH) and SG protein component granule localization in intact cells, in cells where the plasma membrane was selectively solubilized by addition of the sterol-binding detergent digitonin, and in cells where the ER membrane was detergent solubilized following digitonin-dependent extraction of the cytosol. Notably, SGs retained their canonical perinuclear enrichment following disruption of the plasma membrane and extraction of the cytoplasm but were lost upon detergent solubilization of the ER membrane (and other organelle membrane systems). These data are consistent with a role for the ER as the predominant subcellular site for SG biogenesis, either via direct interaction with ER resident membrane proteins and/or via assembly on ER-anchored mRNAs [46, 68, 70–73]. Distinguishing between these two scenarios will require experimental approaches that can distinguish direct vs. mRNA-tethered membrane assembly models. Importantly, and although only a limited number of mRNAs were examined, robust SG recruitment was observed for ER-targeted mRNAs, in particular mRNAs such as GRP94 that display very high (near unity) steady-state enrichments on the ER [46]. Also of interest, mRNAs such as nucleolin/NCL, which encodes a soluble cytoplasmic/nucleoplasmic protein, and which are not enriched on the ER, also displayed clear recruitment to ER-associated/ER proximal SGS [46].

4. Is Gene Transcriptional Status Linked to SG mRNA Recruitment Efficiency?

An understanding of the biological function(s) of SGs will be aided by insights into how and why mRNAs are selected for recruitment to SGs. Current estimates indicating that ca. 10% of the total cellular mRNA pool is recruited to SGs in response to the oxidative stressor sodium arsenite and that for most genes only a small fraction of their transcripts undergo recruitment into SGs [44, 74]. Yet, stressors that activate the ISR, such as sodium arsenite, provoke a uniform inhibition of translation initiation and polyribosome breakdown [26, 28]. If the inhibition of translation initiation was alone necessary and sufficient for mRNA recruitment into SGs it would be expected that the SG mRNA transcriptome composition would mirror bulk cellular mRNAs in both composition and abundance. An investigation of the SG transcriptome, coupled with smFISH studies of SG recruitment efficiency, revealed however that the criteria by which mRNAs are selected for SG recruitment are complex [44]. Particularly notable were the findings that mRNA recruitment efficiencies varied from near zero to near uniformity (<1% to > 95%) and that recruitment efficiencies were correlated with transcript length and translational efficiencies, where transcript length was positively correlated and translational efficiency negatively correlated with SG recruitment efficiency [44]. There are numerous possible explanations for these findings, ranging from limiting SG nucleating and/or assembly factors concentrations to variations in local mRNA concentrations and/or mRNA identities, and others (see [53] for a discussion of the experimental challenges confronting the understanding of SG mRNA recruitment selectivity). It should also be considered that recruitment of mRNAs into nascent SGs is not a broadly stochastic process but rather that the fraction of mRNAs undergoing SG recruitment can be distinguished from the SG recruitment-refractory fraction. In this view, a distinct subset of mRNAs undergoes efficient recruitment and most mRNAs, regardless of translation status, are largely excluded. This view came under consideration when examining ER-associated SG formation during the UPR where UPR activation yields an inhibition of translation initiation via PERK and transcriptional upregulation, via IRE1 and ATF6 [75–78]. Here it was noted that ER-targeted mRNAs undergoing transcriptional upregulation in response to UPR activation, e.g., GRP94 and CCN2, were efficiently recruited to SGs whereas ER-targeted mRNAs that did not undergo UPR-dependent transcriptional upregulation, e.g., B2M, did not [46]. Intriguingly, UPR activation in the presence of the transcription inhibitors actinomycin D or triptolide abrogated ER-associated SG formation without impacting PERK-mediated translational repression or UPR activation (e.g., IRE1-dependent splicing activation of the XBP1 UPR transcription factor) [46]. These findings are consistent with a model where newly transcribed and processed mRNAs, exported under conditions of suppressed translation initiation, are preferred substrates for SG recruitment. This model is also consistent with prior studies of viral regulation of SG biogenesis, notably cricket paralysis virus (CrPV), where CrPV elicits a rapid inhibition of host protein synthesis, yet SG formation is suppressed [79, 80]. Interestingly, CrPV infection compromises SG formation in response to sodium arsenite or pateamine A addition [80]. Subsequent work demonstrated that the CrPV-dependent suppression of SG formation was mediated by cricket paralysis virus 1A protein (CrPV-1A), a multifunctional viral protein which is nuclear-localized and can function as a transcriptional repressor [80]. Further evidence for the importance of de novo transcription in SG biology was provided by Bounedjah et al., who reported that actinomycin D prevented arsenite-induced SG formation [81], though these authors emphasize enhanced mRNA interactions with aggregation-prone RBPs or misfolded proteins as the mechanism of actinomycin D suppression of SG formation. Nonetheless, and with the caveat of limited supportive evidence, it appears that de novo transcription plays a significant role in SG biology and thus that gene transcriptional status may be linked to SG mRNA recruitment efficiency. This is further discussed in the context of a working model describing ER function in SG biogenesis, presented below.

5. A Model for ER Function in SG Biogenesis.

The experimental observations noted above support a role for the ER in the processes that regulate RNP granule assembly. The question can be posed as to why SG formation might occur on the ER rather than or in addition to other endomembrane organelles, or the cytoplasm, if future experimental evidence indicates that SGs assemble largely on the ER. A simple answer may be that the ER is a primary site of mRNA translation in the cell, with the entire mRNA transcriptome being at least partially localized and translated on ER-associated ribosomes, with the ER functioning as a subcellular region of high effective mRNA, ribosome, translation factor, and RBP concentrations. To explain a proposed link between gene transcriptional state and SG recruitment efficiency, it is useful to first note the spatial organization of the ER in the cell and that the ER is physically continuous with the outer nuclear envelope (Fig. 2). A spatial consequence of the continuous apposition of the ER with the outer nuclear envelope is that newly exported mRNAs are likely to encounter ER-associated translation machinery coincident with nuclear export. An immediate coupling of nuclear export with mRNA translation was previously noted in ultrastructural studies of Balbiani ring RNP export [82, 83], where it was demonstrated that mRNAs are exported 5’-3’, with co-export association of ribosomes on the 5’ of the Balbiani ring mRNA as it emerged from the nuclear pore [83].

The early/pioneer rounds of translation occurring on newly exported mRNAs, as part of mRNA quality control processes such as nonsense codon scanning, involve an extensive remodeling of RNP composition, where nuclear cap and polyA tail binding proteins are exchanged, exon-junction complex proteins displaced as are numerous prion/aggregation domain-bearing nuclear RNA binding proteins, which are then imported back into the nucleus by the nuclear import machinery [84–88]. Under stress conditions where translation initiation is inhibited, newly exported mRNAs would be compromised in their ability to undergo such RBP remodeling. Consequently, this cohort of mRNPs would retain a nuclear RBP “fingerprint” for a relatively extended time period and thus display a higher probability of recruitment to nascent SGs. Consistent with this model, evidence for the ER as a site of NMD has been reported [86], and it has also been reported that UPR activation and the consequent inhibition of translation initiation suppresses NMD activity [89]. Also relevant, the ER has been demonstrated to serve as the site of miRNA silencing, known to occur on newly exported or “adolescent” mRNAs [90, 91]. As discussed above in the context of Balbiani ring mRNA export, newly exported mRNAs undergo a bevy of quality control and regulatory interactions that occur coincident with RBP remodeling and likely include secondary structural alterations in the mRNA itself. Because these processes are largely localized to the ER membrane, newly exported mRNAs can be viewed as experiencing structural state changes (RBP composition and secondary structure) while in association with the ER and engaged in their initial or pioneer round(s) of translation. We hypothesize that such state changes can be determinants in ER-associated SG biogenesis. As noted above, such state changes could reflect engagement of the newly exported mRNP with ER-associated ribosomes/ribosomal subunits, translation factors, DEAD box helicases, NMD machinery, etc., all of which contribute to post-nuclear export RBP remodeling. During stress-induced inhibition of translation initiation, newly exported mRNAs may engage ER-bound ribosomal small subunits, translation initiation factors (yielding dead-end mRNA-initiation site scanning complexes) and/or ER integral membrane RBPs and in this non- or weakly translated state undergo recruitment into SGs.

If mRNA transcription, export, and SG recruitment are mechanistically linked, a number of experimentally testable predictions can be inferred. One, mRNA export rate should positively correlate with SG recruitment efficiency; if newly exported mRNAs are the primary substrates for SG biogenesis, then mRNA recruitment efficiency would be expected to correlate with transcription/export rates. Although this prediction is challenged by the finding that heat shock and other stress conditions suppress nucleocytoplasmic transport in budding yeast, in the absence of a quantitative understanding of the magnitude and gene selectivity of this response, it may be that mRNA export levels sufficient to fuel SG biogenesis can occur during evoked stress [92, 93]. Two, newly exported mRNAs display biased localization to the ER. If the ER represents the primary site of mRNA translational quality control, it follows that newly exported mRNAs, regardless of their final subcellular localization destination, are predominately localized to the ER. Extending from this prediction, it would also be expected that during cell stress/suppression of translation initiation, newly exported mRNAs would be strongly biased to an ER distribution. Three, mRNA susceptibility to SG recruitment declines as a function of time post nuclear export such that SG is biased to “adolescent” mRNAs. Under homeostatic conditions, in contrast, newly exported mRNAs would be efficiently translated, their RBP compositions rapidly remodeled, and consequently they would be rendered refractory to SG recruitment. Four, the three predictions note above would be influenced by the intrinsic translational efficiency and length of a given mRNA, with enhanced intrinsic translational efficiencies negatively correlating and CDS length positively correlating with SG recruitment efficiencies, as previously reported by Khong et al. [44]. We suggest that increasing CDS length would be positively correlated with bound nuclear RBP abundance and consequently a higher likelihood for the multivalent, avidity-based interactions that are accepted to support SG recruitment. Intrinsic translational efficiencies as well as translational efficiency during elevated phospho-eIF2α:eIF2α ratios would also govern SG recruitment efficiencies, also as reported by Khong et al. [44]. Combined, this model predicts that mRNA SG recruitment efficiency would be influenced by transcription/export rates and the translational efficiency in the presence of elevated phospho-eIF2α:eIF2α, which determine nuclear RBP remodeling kinetics that are unique to newly exported mRNAs. These predictions, introduced in an earlier publication [44], are schematically illustrated in Fig. 2, and focus on the immediate localization and translational recruitment dynamics of newly exported mRNAs as primary regulators of mRNA recruitment into ER-associated SGs.

Summary:

Here we propose that the ER serves as the primary site of SG biogenesis and highlight the biochemical processes that accompany nuclear export as key determinants for the provision of suitable mRNAs and mRNA structural state conditions for SG biogenesis. This view is speculative though supported by recent work demonstrating that SGs can form in close apposition to the ER and that this process is gene-selective [46], recent optical imaging data demonstrating that PBs, which exhibit functional crosstalk with SGs, interact with the ER and their structural dynamics can be regulated through such interactions [61], and the recent finding that in fungi, an RBP known to promote liquid-liquid phase separation of mRNAs, is predominately ER-bound [63, 64]. A key question in need of investigation is if an ER-centric site for SG biogenesis reflects the dimensionality reduction and kinetic interaction enhancements that accompany surface localization of reactions or, alternatively, if ER-centric SG biogenesis reflects the subcellular spatial biology of newly exported mRNAs, where the state changes that accompany the transition from a nuclear environment, where transcription, capping, polyadenylation, splicing, and packaging for export are primary, to a cytosolic environment, where translation, localization, and stability regulation are primary. These two scenarios need not be mutually exclusive. Regardless, a better understanding of the spatial organization of the processes that accompany nuclear export of newly transcribed mRNAs and their immediate translation, quality control scanning, and localization is likely to yield significant new insights into PB and SG biology, in particular their early biogenesis.