Translating across kingdoms: target of rapamycin promotes protein synthesis through conserved and divergent pathways in plants

Recent advances enabled by cutting-edge genomics and proteomics have illuminated how plants coordinate protein synthesis with nutrient availability through a conserved signaling hub, the target of rapamycin (TOR) kinase.

Abstract

mRNA translation is the growth rate-limiting step in genome expression. Target of rapamycin (TOR) evolved a central regulatory role in eukaryotes as a signaling hub that monitors nutrient availability to maintain homeostasis and promote growth, largely by increasing the rate of translation initiation and protein synthesis. The dynamic pathways engaged by TOR to regulate translation remain debated even in well-studied yeast and mammalian models, however, despite decades of intense investigation. Recent studies have firmly established that TOR also regulates mRNA translation in plants through conserved mechanisms, such as the TOR–LARP1–5ʹTOP signaling axis, and through pathways specific to plants. Here, we review recent advances in our understanding of the regulation of mRNA translation in plants by TOR.

Introduction

Target of rapamycin (TOR) coordinates eukaryotic metabolism by integrating diverse upstream cues, including nutrient availability, hormones, and environmental conditions, and transducing these signals through downstream effectors that promote growth and development when conditions are favorable (Blenis, 2017; Shi et al., 2018; Valvezan and Manning, 2019; Liu and Sabatini, 2020). TOR is strictly required for homeostasis (Valvezan and Manning, 2019): mutations that hyperactivate TOR permit uncontrolled growth and metabolic crises (including human cancers) (Chresta et al., 2010; Zaytseva et al., 2012; Grabiner et al., 2014), whereas mutations that reduce TOR activity disrupt proper development and slow or arrest growth. Biomedical researchers have intensively studied TOR since its discovery almost 30 years ago because TOR dysregulation causes or contributes to diverse diseases and age-related disorders (Laplante and Sabatini, 2012; Lamming et al., 2013; Saxton and Sabatini, 2017). TOR has gained recent attention from plant biologists for its manifold roles in fundamental processes unique to plant biology, including seedling establishment (Xiong et al., 2013), the sink to source transition (Brunkard et al., 2020), flowering time (Moreau et al., 2012), photosynthesis and related light signaling (Li et al., 2017; Riegler et al., 2021; Mallén-Ponce et al., 2022), and leaf shape and patterning (Cao et al., 2019), to name only a few.

TOR is a remarkably conserved atypical serine/threonine protein kinase that phosphorylates a suite of signaling and metabolic proteins in the cytosol. Many of these substrates are themselves protein kinases, including ribosomal protein eS6 serine/threonine kinases (S6Ks) (Chung et al., 1992; Price et al., 1992; Fox et al., 1998; Dufner and Thomas, 1999; Xiong and Sheen, 2012) and the dual-specificity tyrosine/serine/threonine kinase YET ANOTHER KINASE 1 (YAK1) (Garrett et al., 1991; Hartley et al., 1994; Martin et al., 2004; Schmelzle et al., 2004; Barrada et al., 2019; Forzani et al., 2019), which act in kinase signal transduction cascades to regulate gene expression and protein activity. Early studies in yeast and animal models demonstrated that most of the effect of TOR on growth is mediated by the regulation of mRNA translation (Thomas and Hall, 1997). Amino acids and nucleotides activate TOR, which then selectively induces ribosome biogenesis to globally increase the rate of protein synthesis (Iadevaia et al., 2014; Emmanuel et al., 2017; Hoxhaj et al., 2017; Valvezan et al., 2017; Wolfson and Sabatini, 2017; Cao et al., 2019; Schaufelberger et al., 2019; O’Leary et al., 2020; Busche et al., 2021; Liu et al., 2021; Melick and Jewell, 2020; Vellai, 2021; Lutt and Brunkard, 2022). This simple paradigm is conserved in all eukaryotes studied to date (Brunkard, 2020), although the actual regulatory mechanisms are far more complicated than the straightforward three-step model (amino acids and nucleotides→TOR→translation) would suggest. Here, we will first provide a brief ‘primer’ on the eukaryotic mechanism of translation, followed by a summary of the current understanding of how TOR regulates translation in mammalian models. Then, we will focus on breakthrough reports in plant models highlighting how TOR modifies the translational apparatus at multiple regulatory steps, and conclude with an in-depth discussion of the plant TOR–LARP1–5ʹTOP signaling axis.

Translating mRNA: an overview

mRNA translation consists of three stages: initiation, elongation, and termination (Kapp and Lorsch, 2004; Jackson et al., 2010; Browning and Bailey-Serres, 2015; Sokabe and Fraser, 2019) (Fig. 1). The translation machinery is composed of proteins that are well conserved across eukaryotes, including eukaryotic initiation factors (eIFs), eukaryotic elongation factors (eEFs), and eukaryotic release factors (eRFs). The first step of translation initiation is mRNP formation via the interaction of mRNA with a cap-binding complex called eIF4F, which is composed of three subunits (eIF4G, eIF4E, and eIF4A), at the 5ʹ end of the mRNA and a poly(A)-binding protein (PABP) at the 3ʹ end of the mRNA. This is followed by the addition of two more initiation factors, eIF4A and eIF4B, before recruiting the 43S pre-initiation complex (43S PIC). The 43S PIC forms from the association of the small ribosomal subunit (40S) and several initiation factors (eIF1, eIF1A, eIF3, and eIF5) and, subsequently, the ternary complex (eIF2, Met-tRNAi, and GTP). All of these complexes together with the mRNA are called the open- conformation 48S complex, which scans the mRNA in the 5ʹ to 3ʹ direction with the initiation tRNA (Met-tRNAi) until Met-tRNAi recognizes the start codon (AUG). Once the start codon is selected, the 48S complex transitions to a closed conformation, ejects some of the initiation factors, recruits the eIF6-bound 60S (large) ribosomal subunit to assemble the 80S ribosome with the support of the eIF5B GTPase, and eventually ejects the remaining initiation factors during the first rounds of translation elongation. Briefly, elongation proceeds by ribosomes scanning tRNAs with eEF1A, transferring peptides when a tRNA that matches the mRNA codon is recognized, and then moving the ribosome along the mRNA with the support of GTPase eEF2. This cycle repeats until the ribosome encounters one of the three stop codons (UAG, UAA, or UGA), which triggers the final stage, translation termination. eRF1 recognizes the stop codon and stimulates peptide release and ribosome dissociation, supported by the GTPase eRF3 and ATP-binding protein ABCE1. Diverse quality control systems monitor translation at all of these steps to ensure efficient, high-fidelity protein synthesis (Brunkard and Baker, 2018; Schuller and Green, 2018; Joazeiro, 2019).

Fig. 1.

TOR regulates the expression and activity of the plant translation apparatus. Translation initiates by loading mRNAs with eIF4A/4B, eIF4F (eIF4E/4G; in plants, this may instead be eIF4isoF), and PABP to form mRNPs. eIF4F/eIFiso4F/PABP-bound mRNAs recruit the 43S pre-initiation complex, composed of eIF1/1A, eIF2/2B, eIF3, eIF5, and the 40S small ribosomal subunit, resulting in the scanning 48S complex. Once the start codon is recognized, 60S joins and the remaining initiation factors are released, allowing the functional 80S ribosome to proceed with translation elongation. Translation terminates when eRF1/eRF3 recognize a stop codon and trigger release of the polypeptide and ribosome splitting, allowing the 40S and 60S subunits to be recycled for future translation. LARP1 regulates the translation of 5ʹTOP mRNAs in response to TOR activation. Each subunit is labeled to indicate whether some of its components are transcriptionally induced by TOR (green triangle with ‘T’), translationally regulated by TOR–LARP1 signaling (‘5ʹTOP’), or differentially phosphorylated in response to TOR activation (blue circle with ‘P’ indicates TOR-induced phosphorylation, red circle with ‘P’ indicates TOR-repressed phosphorylation).

TOR engages defined pathways to regulate translation in mammals

In yeast and animals, TOR controls protein synthesis at multiple regulatory steps, especially the rate-limiting step, translation initiation (Shah et al., 2013), to increase the translation of specific mRNAs and broadly up-regulate the global rate of protein synthesis (Ma and Blenis, 2009). TOR directly or indirectly phosphorylates a subset of proteins involved in translation initiation (Jiang et al., 2016), including eIF4B (Holz et al., 2005; Shahbazian et al., 2006), eIF4G (Raught, 2000), eIF4E-binding proteins (4EBPs) (Pause et al., 1994; Haghighat et al., 1995; Beretta et al., 1996; Hara et al., 1998), and La-related protein 1 (LARP1) (Kang et al., 2013; Fonseca et al., 2015; Berman et al., 2021; Jia et al., 2021). Each of these phosphorylation events impacts the translation initiation of different subsets of transcripts, typically based on the sequence or structure of their 5ʹ leaders (Leppek et al., 2018). eIF4B phosphorylation specifically enhances translation of eIF4A/4B-sensitive transcripts, which are typically long mRNAs that begin with highly structured 5ʹ leaders (Sen et al., 2016). eIF4F assembly is promoted by eIF4G phosphorylation, which stimulates its association with eIF4E, and 4EBP phosphorylation, which prevents 4EBPs from sequestering eIF4E. The transcripts most sensitive to eIF4G phosphorylation tend to be short, highly expressed, and unstructured (Gandin et al., 2016; Sen et al., 2016). Thus, eIF4G and eIF4A/4B phosphorylation regulate the translation initiation of distinct mRNA populations. LARP1 binds specifically to the 5ʹ cap of mRNAs with a ‘tract of oligopyrimidines’ (5ʹTOP) motif and represses their translation (Philippe et al., 2020); LARP1 phosphorylation stimulates its release from the 5ʹ cap, thereby permitting translation of 5ʹTOP mRNAs (Fonseca et al., 2015; Hong et al., 2017; Lahr et al., 2017; Philippe et al., 2018, 2020; Cassidy et al., 2019; Scarpin et al., 2020; Jia et al., 2021). Many 5ʹTOP mRNAs are also short, highly expressed, and unstructured, so the set of transcripts regulated by LARP1 and eIF4F overlaps considerably, which has led to some debate over the relative significance of these two mechanisms (Thoreen et al., 2012; Philippe et al., 2020; Berman et al., 2021).

For many of its functions in translation, TOR acts indirectly through kinase signaling cascades. Ribosomal protein S6Ks are directly phosphorylated by TOR at a conserved activating threonine (T389 in humans, T449 in Arabidopsis) and then phosphorylate diverse protein substrates, including some of those described above (e.g. eIF4B and LARP1). S6Ks also phosphorylate C-terminal residues of ribosomal protein eS6, which may influence translation (Ruvinsky et al., 2005; Chen et al., 2018; Mancera-Martínez et al., 2021), although the functional significance of eS6 phosphorylation remains contested. Note that many of the TOR-dependent kinase signaling cascades integrate other upstream signaling cues; for example, full S6K activity in humans and probably also in plants requires additional post-translational modifications, including phosphorylation by PHOSPHOINOSITIDE-DEPENDENT PROTEIN KINASE 1 (PDK1) at another site (T252 in humans, S290 in Arabidopsis). Indeed, several ribosomal proteins are differentially phosphorylated in response to TOR activity, but the molecular impact of ribosomal protein phosphorylation is relatively understudied (Jiang et al., 2016). TOR–S6K signaling also regulates translation elongation: S6Ks activate eEF2 by phosphorylating the eEF2 kinase (Wang, 2001), accelerating the rate of protein synthesis (Ryazanov et al., 1988; Proud, 2019). Unlike translation initiation, it is not established whether specific mRNAs are sensitive to eEF2 phosphorylation, but it has been proposed that transcripts already prone to inefficient translation and ribosome collisions due to suboptimal codon arrangements or mRNA secondary structures might be especially responsive to the TOR–S6K–eEF2K–eEF2 signaling axis.

TOR increases the overall rate of protein synthesis by stimulating ribosome biogenesis. In mammals, all ribosomal proteins are encoded by 5ʹTOP mRNAs and translationally stimulated by TOR–LARP1 signaling (Shama and Meyuhas, 1996). LARP1 was first identified in Drosophila (Chauvet et al., 2000), where it was later shown to interact with PABPs, hinting at its role in regulating mRNA translation (Blagden et al., 2009; Burrows et al., 2010). LARP1 orthologs were then investigated in various model systems, including Caenorhabditis elegans (Nykamp et al., 2008), Arabidopsis thaliana (Merret et al., 2013), and mammals (Aoki et al., 2013; Kang et al., 2013), with a general consensus that LARP1 participates in RNA metabolism but no clearly defined molecular mechanism. A proteomic screen for TOR-sensitive 5ʹ cap-binding proteins first revealed that LARP1 responds to TOR activity (Tcherkezian et al., 2014), although early studies debated whether LARP1 promotes or represses 5ʹTOP mRNA translation. These debates stimulated considerable investigation of TOR–LARP1 signaling, revealing that LARP1 may have additional roles in translation efficiency and mRNA stability, but generally supporting the current consensus model that LARP1 binds to 5ʹTOP mRNAs and specifically represses their translation initiation when TOR is inactive (Berman et al., 2021). TOR therefore acts through the TOR–LARP1–5ʹTOP signaling axis to increase expression of mammalian ribosomal protein mRNAs, which effectively elevates the translation efficiency of all transcripts.

Phosphoproteomics illuminate how TOR regulates translation in plants

Foundational studies of TOR in Arabidopsis established that TOR also promotes translation in plants (Deprost et al., 2007; Chen et al., 2018), but relatively little is known about the underlying mechanisms. Many of the translation-related proteins downstream from TOR are conserved in plants, including eIF4G, S6K, and LARP1. Others have diverged considerably, including eIF4B (Browning et al., 1989), which has evolved significant differences across eukaryotic lineages, and eIFiso4G, a plant-specific isoform of eIF4G encoded by a paralogous gene (Browning et al., 1987, 1992; Castellano and Merchante, 2021). There are no conserved plant orthologs of 4EBPs (Brunkard, 2020), although several other proteins have been shown to interact with eIF4E through a 4EBP-like eIF4E-binding motif, including CONSERVED BINDING OF eIF4E 1 (CBE1) (Patrick et al., 2018) and CERES (Toribio et al., 2019). Mechanistically, of these phosphoproteins, only S6K and LARP1 have been shown to regulate translation downstream from TOR in plants (Schepetilnikov et al., 2013; Scarpin et al., 2020). TOR–S6K signaling promotes translation reinitiation of mRNAs that include short upstream ORFs (uORFs) within their 5ʹ leaders (Schepetilnikov et al., 2013, 2017; Mancera-Martínez et al., 2021). Inhibiting TOR acutely represses translation of mRNAs that harbor uORFs in a pathway dependent on phosphorylation of eIF3h by S6K (Schepetilnikov et al., 2011, 2013). TOR–LARP1–5ʹTOP signaling is conserved in plants and ­animals, specifically regulating the translation of 5ʹTOP mRNAs involved in ribosome biogenesis and other processes (see next section, ‘On TOP of translation’) (Scarpin et al., 2020).

To establish how TOR coordinates metabolism in plants, Van Leene et al. (2019) and Scarpin et al. (2020) conducted large-scale phosphoproteomic screens for substrates and downstream targets of TOR, revealing ~160 TOR-sensitive phosphoproteins in plants (Van Leene et al., 2019; Scarpin et al., 2020). Translation initiation factors and cytosolic ribosomal proteins are highly enriched in these data, implying that TOR could fine-tune many steps of translation initiation in plants (Fig. 1). Several proteins involved in mRNP loading, the first step of translation initiation, are phosphorylated when TOR is active, including 5ʹ cap-binding protein LARP1, eIF4E-binding protein CBE1, eIF4A, eIF4B, and PABP. In contrast, eIF4G is phosphorylated only when TOR is inactive, suggesting that an antagonistic kinase signaling pathway may also regulate translation initiation. In the 43S and 48S PICs, TOR promotes phosphorylation of subunits of the GTPase eIF2 and its guanine exchange factor, eIF2B, which are critical for accurate selection of start codons during scanning in other eukaryotes. Several ribosomal proteins in the 40S and 60S ribosomal subunits are also differentially phosphorylated in response to TOR, prominently including C-terminal phosphosites of eS6, which is a direct substrate of TOR-activated S6K.

Of these putative TOR effectors, to date, only LARP1 is established to function downstream of TOR in regulating translation (Scarpin et al., 2020). Nonetheless, the TOR-sensitive translatome includes many transcripts that are still translationally regulated by TOR in larp1 mutants, demonstrating that other factors must be involved in TOR responses. These LARP1-independent impacts on translation efficiency could be mediated by the other translationally regulated phosphoprotein effectors of TOR. In parallel, TOR up-regulates the transcription and/or stability of myriad mRNAs involved in translation, including transcripts that encode every ribosomal protein subunit, almost all initiation and elongation factors, and various accessory proteins. The TOR-sensitive proteome further revealed that TOR is required to maintain high levels of eIF3, eEF1B, and ribosomal proteins (Scarpin et al., 2020), probably through control of synthesis and degradation rates (e.g. by suppressing ribosomal autophagy, or ‘ribophagy’; Hillwig et al., 2011; Wyant et al., 2018). Therefore, TOR also fine-tunes translation efficiency through downstream effects on RNA and protein metabolism, and comprehensive investigations of these various steps will be needed to fully appreciate how TOR regulates translation in plants.

On TOP of translation: new and conserved targets of TOR–LARP1–5ʹTOP signaling

We recently proposed that the TOR–LARP1–5ʹTOP signaling axis evolved early in eukaryotes and established several 5ʹTOP mRNAs that are conserved in humans and plants (Scarpin et al., 2020). These ‘core’ eukaryotic 5ʹTOP mRNAs encode proteins involved in ribosome biogenesis, mRNA translation, mRNA splicing, cell cycle progression, and vesicle trafficking. TOR–LARP1–5ʹTOP signaling also regulates translation of several mRNAs specific to plants, including genes involved in cell wall synthesis and modification, plastid biogenesis, and developmental patterning.

Discovering regulatory features of 5ʹ leaders in plants will require extensive annotation of transcription start sites (TSSs), which are not trivial to identify (Leppek et al., 2018). TSSs vary stochastically within cells, across different cell types, and in response to dynamic environmental and physiological cues (Kurihara et al., 2018). Moreover, standard RNA-Seq approaches are significantly biased against 5ʹ ends and inaccurately resolve TSSs. Methods including 5ʹCAP-Seq (TSS-Seq), which involves ‘trapping’ the 5ʹ cap and sequencing only from the 5ʹ end of capped mRNAs, have become popular in human genomics to precisely define 5ʹ leader sequences (Pelechano et al., 2016). Several groups have recently used 5ʹCAP-Seq and related approaches to begin defining TSSs in plant genomes (Kindgren et al., 2018; Nielsen et al., 2019), which will be crucial to understanding the molecular regulatory mechanisms that coordinate plant mRNA translation.

To illustrate this point, we expanded our search for 5ʹTOP motifs using recently published TSS sequences identified by TSS-Seq in Arabidopsis (Nielsen et al., 2019), which includes tissues and growth conditions we had not analyzed in our previous report. 5ʹTOP motifs are quantitatively defined by ‘TOPscores’, which measure the number of uninterrupted pyrimidines in every sequenced TSS for a gene (Philippe et al., 2020). The vast majority of genes (>87%) have TOPscores between 0 and 1, indicating that their transcripts start with purines or a mixture of pyrimidines and purines. The new annotations revealed several more ‘core’ eukaryotic 5ʹTOP mRNAs (Table 1), including the nucleosome assembly protein NAP1, another subunit of the eukaryotic elongation factor 1B (eEF1Bγ1), and several cytosolic ribosomal proteins (eS6, eS7, uS11, eS24, eL13, and uL30). Translation initiation factors are also common targets of TOR–LARP1–5ʹTOP signaling, including eIF3 subunits (eIF3m in Arabidopsis and eIF3a,e,f,h in humans). With the addition of NAP1, eIF3, and more cytosolic ribosomal proteins, it is now clear that virtually all categories of the ‘canonical’ human 5ʹTOP mRNAs are conserved targets of TOR–LARP1 signaling in both plants and animals.

Table 1.

Newly identified 5ʹTOP mRNAs translationally regulated by TOR–LARP1 signaling in plants.

Locus ID	Gene	TOPscore	5ʹTOP motif	Wild-type ΔTE (T2/mock)	larp1 ΔTE (T2/mock)
Conserved eukaryotic 5ʹTOP mRNAs
At3g02200	eIF3m	4.5	CUCUCU	-0.6	0.0
At1g09640	eEF1Bγ1	4.3	CUCUCU	-0.6	0.0
At4g26110	NAP1	1.5	CUUGUC	-1.1	-0.1
Cytosolic ribosomal protein 5ʹTOP mRNAs
At3g13580	uL30	5.5	CUUUUU	-0.7	0.0
At5g28060	eS24	3.3	CUUCCA	-0.8	-0.2
At2g36160	uS11	3.2	CUCUCU	-0.8	-0.1
At4g34670	eS1	2.7	CUCUCA	-0.9	-0.1
At5g23900	eL13	2.5	CUCUGG	-0.9	-0.1
At3g02560	eS7	2.5	CUCCCU	-0.8	0.0
At5g10360	eS6	1.7	CCUAAA	-0.7	-0.1
Select plant-specific 5ʹTOP mRNAs
At1g79880	LA2	6.4	CUCUCU	-0.8	0.2
At2g36400	GRF3	5.4	CUCUCU	-1.1	0.6
At5g59440	ZEUS1	6.3	CUCUUU	-1.4	0.7

ΔTE (T2/mock) indicates log₂(fold change in translation efficiency) after 2 h of TOR inhibition, as previously described (Scarpin et al., 2020).

Beyond the core 5ʹTOP mRNAs, this analysis also revealed several more 5ʹTOP mRNAs involved in TOR-regulated metabolism (Table 1). To highlight only a few, these include the growth-promoting transcription factor GRF3 (van der Knaap et al., 2000; Kim et al., 2003), the tRNA maturation factor La2 required for efficient translation (Fleurdépine et al., 2007), and the thymidylate kinase ZEUS1 that contributes to pyrimidine biosynthesis and DNA replication during cell division (Ronceret et al., 2008; Xiong and Sheen, 2013; Xiong et al., 2013; Busche et al., 2021). These examples illustrate how TOR could fine-tune translation efficiency through the LARP1 signaling axis to coordinate plant growth by impacting multiple metabolic and signaling pathways.

To confidently define 5ʹTOP mRNAs, we used stringent criteria, including a requirement that the mRNA is translationally promoted by TOR in a LARP1-dependent mechanism. These criteria excluded, therefore, any mRNAs that were expressed at a low level and therefore not quantified in the TOR- and LARP1-sensitive translatomes. There are several exciting candidates among these mRNAs with clear 5ʹTOP motifs, however, including multiple genes that ­participate in metabolic signaling (Table 2), which may warrant future investigation. For example, several FLZ (FCS-like Zinc Finger) proteins are encoded by mRNAs with 5ʹTOP motifs, including FLZ3, FLZ4, FLZ5, FLZ6, and FLZ7. FLZ proteins interact with and negatively regulate SnRK1, a protein kinase that is activated when sugars are low (i.e. starvation conditions, similar to its homolog AMP-activated kinase in mammals) (Baena-González et al., 2007; Robaglia et al., 2012; Nukarinen et al., 2016; Baena-González and Hanson, 2017; Li et al., 2021). SnRK1 broadly antagonizes TOR signaling, and TOR activity is attenuated in some flz mutants, supporting a model whereby FLZ proteins promote TOR signaling by repressing SnRK1 (Jamsheer et al., 2018a, 2018b, 2019). Given the striking 5ʹTOP motifs found on FLZ mRNAs, it is likely that TOR–LARP1–5ʹTOP fine-tunes FLZ expression at the translational level, inducing FLZ expression and suppressing SnRK1 under favorable conditions for growth.

Table 2.

FLZ–SnRK1–bZIP signaling axis transcripts start with 5ʹTOP motifs (Scarpin et al., 2020)

Locus ID	Gene	TOPscore	5ʹTOP motif
FCS-like zinc finger 5ʹTOP mRNAs
At1g22160	FLZ5	13.7	CUCUUU
At1g78020	FLZ6	13.3	CCCCUC
At2g44670	FLZ3	10.9	CUCCUC
At4g39795	FLZ7	5.4	CCCUCU
At5g65040	FLZ4	3.4	CCCCUU
bZIP transcription factor 5ʹTOP mRNAs
At4g34590	bZIP11	14.3	CUCUUC
At1g75390	bZIP44	6.9	CUUCCC
At3g62420	bZIP53	6.3	UUCUUU
At2g18160	bZIP2	6.1	CUCUUU
At5g49450	bZIP1	2.6	CUCCCA

Among the top candidate 5ʹTOP mRNAs, the S1 subfamily of bZIP transcription factors stand out. S1 bZIP transcription factors are encoded by a paralogous group of five genes in Arabidopsis (bZIP1, bZIP2, bZIP11, bZIP44, and bZIP53) that are well-established regulators of low-energy responses in plants, working downstream of SnRK1 to drive a stress response transcriptional program (Baena-González et al., 2007; Baena-González and Sheen, 2008; Ma et al., 2011; Delatte et al., 2011). Early studies recognized that S1 bZIP expression is translationally regulated by sucrose (Rook et al., 1998); subsequent investigations have argued that this translational regulation is mediated by a conserved short uORF (a ‘conserved peptide uORF/CPuORF’), in the 5ʹ leader of all five genes, which requires translation to reinitiate in order for the main ORF that encodes the bZIP proteins to be expressed (Wiese et al., 2004). Whereas high levels of exogenous sucrose translationally repress bZIP11 expression in a CPuORF-dependent mechanism, TOR attenuates this repression and promotes the translation of bZIP11 transcripts (Schepetilnikov et al., 2013). Strikingly, the bZIP11 mRNA has the eighth highest TOPscore in the entire Arabidopsis genome (~15), indicating that virtually all bZIP11 transcripts start with an uninterrupted stretch of at least six pyrimidines), and the remaining S1 subfamily bZIP mRNAs have scores in the top 95th percentile, strongly suggesting that TOR could regulate their translation through the LARP1–5ʹTOP signaling axis. The discovery of 5ʹTOP motifs on the S1 subfamily of bZIP transcription factors reinforces the complexity of signaling networks that converge on this small family of genes to robustly coordinate starvation responses in plants, and highlights the potential mechanistic insights gained from high-resolution definition of 5ʹ leader sequences genome-wide.

Conclusion

Translational regulation research is experiencing a renaissance in plant biology, propelled by a growing arsenal of tools to globally profile changes in translation efficiency of every transcript in a cell (Ingolia et al., 2009; Juntawong et al., 2014; Hsu et al., 2016; VanInsberghe et al., 2021). Landmark studies have demonstrated, for example, that translational regulation substantially contributes to plant immune systems (Xu et al., 2017a; Fröschel et al., 2021) and abiotic stress responses (Kawaguchi et al., 2004; Branco-Price et al., 2005; Mustroph et al., 2009). Interventions focused on the regulation of translation, illuminated by mechanistic studies in model systems, are effective strategies for improving agricultural crop resilience (Xu et al., 2017b; Bailey-Serres et al., 2019). Therefore, we expect that comprehensively mapping the intersections between the TOR signaling network and the mRNA translation apparatus at a molecular level in plants will contribute to the urgent project of creating a sustainable agricultural future.

Author contributions

MRS and JOB: conceptualization, original draft preparation; JOB: funding acquisition; MRS, CHS, and JOB: data curation, formal analysis, investigation, visualization, review, and editing.

Conflict of interest

The authors declare no conflicts of interest.

Funding

This work was supported by NIH grant DP5-OD023072 to JOB.