Abstract
Quaking RNA binding protein (RBP) isoforms arise from a single Quaking gene, and bind the same RNA motif to regulate splicing, stability, decay, and localization of a large set of RNAs. However, the mechanisms by which the expression of this single gene is controlled to distribute appropriate amounts of each Quaking isoform to regulate such disparate gene expression processes are unknown. Here we explore the separate mechanisms that regulate expression of two isoforms, Quaking-5 (Qk5) and Quaking-6 (Qk6), in mouse muscle cells. We first demonstrate that Qk5 and Qk6 proteins have distinct functions in splicing and translation respectively, enforced primarily through differential subcellular localization. Using isoform-specific depletion, we find both Qk5 and Qk6 act through cis and trans post-transcriptional regulatory mechanisms on their own and each other’s transcripts, creating a network of auto- and cross-regulatory controls. Qk5 has a major role in nuclear RNA stability and splicing, whereas Qk6 acts through translational regulation. In different cell types the cross-regulatory influences discovered here generate a spectrum of Qk5/Qk6 ratios subject to additional cell type and developmental controls. These unexpectedly complex feedback loops underscore the importance of the balance of Qk isoforms, especially where they are key regulators of development and cancer.
Introduction
Sequence-specific RNA-binding proteins (RBPs) regulate nearly every post-transcriptional step of gene expression. Control is exerted by recognition of cis-acting regulatory sequences by the RNA-binding domain (RBD), followed by one or more consequences, including restructuring of the RNA and the protein, altered interactions with other proteins, and the generation of activities that influence the function of RNA polymerase, the spliceosome, the export and decay machinery, and the ribosome (Keene 2007; Fu and Ares 2014). Furthermore, many RBPs control their own levels (autogenous regulation), through the same mechanisms by which they regulate their targets (Wollerton et al. 2004; Lareau et al. 2007; Ni et al. 2007; Damianov and Black 2010; Sun et al. 2010). In many metazoans, multiple RBPs with the same or highly similar RBDs (iso-specific) perform distinct functions but are expressed at the same time in the same cells (Makeyev et al. 2007; Spellman et al. 2007; Yeo et al. 2009; Gehman et al. 2011; Charizanis et al. 2012; Gehman et al. 2012; Singh et al. 2014), suggesting that any functional diversification must be enforced by regulatory mechanisms that prevent crosstalk. These two circumstances, autogenous regulation through binding their own pre-mRNAs and mRNAs, combined with functional diversification using the same RBDs and recognition sequences create a regulatory conundrum: How is the right amount of each distinctly functional iso-specific RBP isoform in a cell established and maintained?
Since all isoforms within a family recognize essentially the same RNA sequence, isoform-specific control is unlikely to be mediated through recognition of distinct cis-acting RNA sequences. Two other mechanisms include 1) isoform-specific localization that restricts function to processes within specific compartments (Caceres et al. 1998; Koizumi et al. 1999; Cazalla et al. 2002; Sanford et al. 2004), and 2) interactions between a specific RBP family member and specific elements of a target core machinery (Roscigno and Garcia-Blanco 1995; Jin et al. 2004; Cheever and Ceman 2009; Sharma et al. 2011) mediated by isoform-specific amino acids. Because many RBPs are encoded by multi-gene families with complex alternative splicing (Underwood et al. 2005), whole menageries of functionally distinct RBP isoforms that recognize the same RNA target can be expressed within single cell types (Lee et al. 2009; Hamada et al. 2013; Lee et al. 2016). How does the cell recognize when each distinctly functional RBP family member is present in an appropriate amount for each process?
To examine this question, we have focused on the single Quaking gene (QKI in human, Qk in mouse), which is required for a broad set of functions in diverse tissues (Ebersole et al. 1996; Darbelli et al. 2016), through its contribution to RNA processing steps including splicing (Hall et al. 2013; van der Veer et al. 2013; Darbelli et al. 2016), localization (Li et al. 2000; Larocque et al. 2002), stability/decay (Li et al. 2000; Larocque et al. 2005; Zearfoss et al. 2011; de Bruin et al. 2016a), translation (Saccomanno L 1999; Zhao et al. 2010), and miRNA processing (Wang et al. 2013; Zong et al. 2014). These processes are regulated by dimeric Qk binding an RNA element that includes ACUAAY and a “half-site” (UAAY) separated by at least one nucleotide (Ryder and Williamson 2004; Galarneau and Richard 2005; Beuck et al. 2012; Teplova et al. 2013). Quaking transcription initiates primarily at a single major site, and in most cell types, three alternatively spliced mRNAs encode three protein isoforms [Quaking-5 (Qk5), Quaking-6 (Qk6), and Quaking-7 (Qk7)] that differ only in the C-terminal tail (Ebersole et al. 1996; Kondo et al. 1999). Although various cell types express different ratios of Qk protein isoforms (Ebersole et al. 1996; Hardy et al. 1996; Hardy 1998; van der Veer et al. 2013; de Bruin et al. 2016a), it is unclear how the relative isoform ratios are maintained in order to support tissue-specific regulated RNA processing but disruption of these ratios leads to developmental defects (Ebersole et al. 1996; Cox et al. 1999), cancer (Sebestyen et al. 2015; de Miguel et al. 2016), and schizophrenia (Aberg et al. 2006).
Many studies of Quaking function have used overexpression of Qk isoforms (Wu et al. 2002; Wang et al. 2013), or depletion strategies or mutant models that do not distinguish which Qk isoform is functional (Hardy et al. 1996; Lu et al. 2003; van der Veer et al. 2013). Here we test specific Qk isoforms for separate functions, and identify in part how the appropriate balance of Qk isoforms is maintained. In mouse myoblasts, Qk5 and Qk6 are the predominantly expressed isoforms and we find that Qk5, but not Qk6, regulates splicing, while Qk6 controls mRNA translation and decay. This functional specificity is mediated by subcellular localization encoded into the unique C-terminal amino acids of these isoforms. Furthermore, the relative expression of Qk protein isoforms is regulated in part, by Qk protein isoforms themselves through both auto- and cross-regulatory influences characteristic of the function of each isoform on its other RNA targets. These findings uncover unexpectedly complex isoform control within a single family of RBPs, and suggest that the relative amounts of each isoform are set in a cell type specific fashion and homeostatically controlled by Qk protein isoform levels themselves.
Results
Qk5 and Qk6 are the predominant isoforms in myoblasts
We analyzed the abundance and localization of Qk isoforms (Fig 1A) in myoblasts and differentiated myotubes (Yaffe and Saxel 1977) using isoform-specific antibodies. Total Qk protein level increases during C2C12 myoblast differentiation ((Hall et al. 2013), Fig 1B), with Qk5 the most abundant, followed by Qk6, and then Qk7, which is barely detectable (Fig 1B). Differentiation leads to myotube fusion ((Yaffe and Saxel 1977) Fig 1C), and each isoform increases proportionately (Fig 1B). Consistent with previous studies in glia (Hardy et al. 1996; Lu et al. 2003), in myoblasts and myotubes Qk5 is primarily nuclear, although some cytoplasmic localization is observed (Fig 1D). In contrast, Qk6 and Qk7 are primarily cytoplasmic, but can be detected in both nuclear and cytoplasmic compartments, with noticeable cell-to-cell heterogeneity (Fig 1D). We conclude that Qk5 and Qk6 are the major Qk isoforms in C2C12 myoblasts, and their overall subcellular localization is similar (although not identical) to that observed in glia.
Qk-dependent alternative splicing regulation requires Qk5, but not Qk6
We previously showed that depletion of all Qk isoforms alters splicing of several hundreds of exons in myoblasts, and that one or more unknown Qk isoform(s) directly promotes inclusion of Capzb exon 9 by binding to downstream intronic ACUAA sequence elements (Hall et al. 2013). Since Qk5 is the most abundant and primarily nuclear Qk isoform in myoblasts, we tested the hypothesis that Qk5 is responsible for splicing regulation. We overexpressed a Myc-tagged cDNA construct for each isoform (Fig S1A), and tested the effect on splicing of the Qk-regulated Capzb exon 9 in the β–globin reporter (pDUP51, Dominski and Kole 1991), which includes ∼200 nucleotides upstream and ∼250 nucleotides downstream of Capzb exon 9 (Dominski and Kole 1991; Hall et al. 2013). To our surprise, overexpression of either predominantly nuclear Myc-Qk5 or nuclear and cytoplasmic Myc-Qk6 (Fig 1E) efficiently activates splicing (Fig 1F).
Previous work has shown that Qk proteins are only stable as a dimer in vivo (Beuck et al. 2012), and that overexpression of Qk isoforms can create ectopic Qk heterodimers (Pilotte et al. 2001). Given this, we hypothesized that overexpression of Myc-Qk6 could drive formation of dimers with Qk5, leading to aberrant Myc-Qk6 nuclear import as part of a Myc-Qk6-Qk5 heterodimer (Pilotte et al. 2001), which could activate exon inclusion. Consistent with this, we observe strong nuclear staining of Myc-Qk6, which appears throughout cells that overexpress it (Fig 1E). To rigorously test Qk6 function in splicing, we used a depletion-replacement strategy in which all endogenous Qk mRNAs were depeleted using an siRNA to a common region, followed by expression of a single isoform carrying translationally silent siRNA resistant mutations (kind gift from Sean Ryder) to create a cellular pool of Qk dominated by a single isoform (Fig S1B). Under the depletion-replacement treatment, inclusion of Capzb reporter exon is specifically increased by replacement with Myc-Qk5, but not with Myc-Qk6 (Fig 2A and 2B). In addition, whereas Myc-Qk5 accumulates efficiently in the nucleus, Myc-Qk6 remains primarily cytoplasmic (Fig 2C). Taken together, we conclude that Qk5 is predominantly nuclear and serves as a splicing factor, but Qk6 is not, suggesting that this isoform regulates translation (Saccomanno L 1999) in myoblasts also and typically does not directly contribute to splicing regulation.
Nuclear presence of the common Qk protein sequence is sufficient for splicing function
The observation that overexpression and nuclear accumulation of Myc-Qk6 activates splicing inappropriately (compare Fig 1E and 1F to 2B and 2C) suggests that nuclear localization of the common (ie non-isoform specific) parts of Qk might be sufficient to activate splicing. The C-terminal 30 amino acids of Qk5 encode a non-canonical NLS (Wu et al. 1999), however other function(s) specifically necessary for splicing could also be contained in this region. To test this, we made two Qk constructs lacking any isoform-specific C-terminal amino acids (i.e., only the common or “body” 311 amino acids shared by all natural Qk isoforms, Fig 1A). One construct encoded an N-terminal HA epitope tag (HA-QkBody), and the other an N-terminal HA tag followed by the canonical SV40 NLS (HA-NLS-QkBody, Fig 2D). Upon expression in myoblasts (Fig S2), these mutant Qk proteins promote inclusion of Capzb exon 9, whether endogenous Qk protein is depleted or not (Fig 2E). This shows that all necessary residues required for Qk splicing functions reside in the body of the Qk protein and not in the Qk5 isoform-specific tail. The HA-QkBody protein is distributed throughout the cell, whereas the HA-NLS-QkBody protein is concentrated in nuclei (Fig 2F), reinforcing the conclusion that nuclear localization of the common regions of the Qk protein is sufficient for splicing function. Furthermore, it is the compartmentalization in the nucleus mediated by the Qk5-specific tail that determines the fate of Qk5 as a splicing factor. The isoform specific tail of Qk6 contributes to its concentration in the cytoplasm and greatly limits its contributions to splicing as compared to the QkBody mutant protein that is localized in both the nucleus and cytoplasm (compare Figs 2C and 2F).
Qk5 is required for expression of Qk6 and Qk7 at the level of RNA accumulation
To refine the understanding of Qk isoform-specific functions obtained from cells expressing single Qk protein isoforms, we selectively targeted individual protein isoforms using isoform-specific siRNAs (siQk5 or siQk6/7; see Fig 3A). Treatment with siQk5, but not siQk6/7 produces the expected alternative splicing changes (Hall et al. 2013) in the endogenous Rai14 cassette exon 11 (loss of repression, Fig 3B) and the Capzb exon 9 splicing reporter (loss of activation, Fig 3C). However, treatment with siQk5 also leads to the unexpected loss of all Qk protein isoforms (Fig 3D, compare to treatment with pan-Qk siRNA), whereas treatment with siQk6/7 leads only to the loss of Qk6 and 7 (Fig 3D and S3A). We are confident that this is due to specific depletion of the Qk5 mRNA by the Qk5-1 siRNA, because two additional Qk5-specific siRNAs (Fig 3A) lead to the same loss of all Qk RNAs (Fig 3E) and proteins (Fig S3B). Depletion of Qk6 and Qk7 proteins by Qk5-specific siRNA is as efficient as that observed with a Qk6/7-specific siRNA (Fig S3A). We interpret these data to mean that cells cannot express Qk6 or Qk7 RNA or protein unless they also or first express Qk5 protein. We conclude that Qk5 is required to promote efficient accumulation of the major mRNAs produced by the Qk gene.
Qk proteins can bind the Qk6 mRNA 3’ UTR (Hafner et al. 2010; Van Nostrand et al. 2016), and phylogenetically conserved Qk sequence motifs are also found in this sequence ((Paz et al. 2014) Fig S3C), however it is unclear whether binding of Qk5 or Qk6, or both occurs at this site, and what the physiological consequences of these binding events might be. To test whether Qk5 contributes to Qk6 protein expression through binding to the Qk6 3’ UTR, we made a reporter gene with the Qk6 3’UTR fused to the renilla luciferase coding region, and co-transfected it with a firefly luciferase control reporter (Fig S3D), and either control (siNT), siQk, or siQk5-1 siRNAs (siQk6/7 was not used in this experiment because it will deplete the Qk6 reporter transcript directly). Treatment with either siQk or siQk5-1 efficiently depletes all Qk protein isoforms (Fig S3D), and we observe no change in reporter protein abundance (Fig S3E). We conclude that Qk proteins are not required for Qk6 3’UTR reporter protein accumulation (Fig S3E). Taken together with Qk5’s nuclear localization, and the observation that depletion of Qk5 mRNA results in loss of Qk6/7 mRNA and protein (Figs 3D, 3E, S3A, and S3B), we suggest that Qk5 regulates Qk6 expression by promoting pre-mRNA accumulation, consistent with its function as a splicing factor (Figs 3B and 3C).
Qk nuclear RNA accumulation requires Qk5 protein
To test the idea that Qk5 promotes overall Qk RNA accumulation (Fig 3E), we performed subcellular fractionation of C2C12 cells treated with siNT, siQk5, or siQk6/7. Because siRNA is effective in the cytoplasm (Sen and Blau 2005), we expect that changes in nuclear RNA result directly or indirectly from reduction in Qk protein, rather than from siRNA treatment. Western blots show efficient knockdown of all Qk protein isoforms by siQk5 in both whole cell extracts (WCE) and nuclear fractions (Fig 4A). In the nuclear fractions Histone H3 is abundant, whereas Gapdh is nearly undetectable (Fig 4A, right), indicating the nuclear fractions are free of cytoplasm. To test if Qk5 is required for nuclear transcript accumulation of Qk isoforms, we extracted RNA from the nuclear fractions and measured Qk RNAs, and as a control 7SK RNA. We find that depletion of Qk5 protein results in significant reduction of both Qk6/7 transcripts, as well as its own transcript in nuclear extracts (Fig 4B). Because this assay measures specific parts of the same pre-mRNA destined for the distinct Qk mRNA isoforms, a simple interpretation is that Qk5 protein is required for stabilization of nuclear Qk pre-mRNA, and that upon loss of this function, all the Qk mRNAs fail to accumulate.
Qk5 promotes accumulation of its own mRNA through splicing
To test whether Qk5 protein promotes accumulation of its own transcript through splicing or polyadenylation, we measured the fraction of Qk5 RNA that remains unspliced upstream of the last isoform-specific 3’ splice site, as well as the fraction of Qk5 RNA that reads through the major Qk5 polyadenylation site relative to total nuclear Qk5 (Fig S4A) after depletion of either Qk5 or Qk6/7 protein (Fig 4A). Consistent with a role for Qk5 in splicing of its own mRNA, unspliced nuclear Qk5 pre-mRNA increases 40% relative to total nuclear Qk5 RNA when Qk5 protein is depleted (p < 0.05; Fig 4C, left side). No significant change in the fraction of transcripts that read through the Qk5 polyadenylation site is observed under Qk5 or Qk6/7 depletion (Fig 4C, right side), suggesting that Qk5 does not regulate polyadenylation of its own mRNA. As a side note, when Qk6 and Qk7 are depleted, the fraction of unspliced nuclear Qk5 RNA decreases (p < 0.01; Fig 4C, left side). This is consistent with observations that Qk6/7 depletion affects other Qk5-regulated splicing events (Capzb exon 9 (p < 0.05; Fig 3C) and Rai14 exon 11 (Fig 3B) because Qk6 normally represses Qk5 translation (see below). Examination of the Qk5 terminal intron and coding sequence 3’end indicates Qk protein binding sites (Van Nostrand et al. 2016), some of which overlap with conserved computationally predicted Qk binding sites ((Paz et al. 2014) Fig S4B). We cloned this region into the β globin splicing reporter pDUP51 ((Dominski and Kole 1991) Fig S4B) to test if Qk5 promotes splicing through this minimal sequence. Depletion of Qk protein with siQk5 (Fig S4C) significantly reduces use of this Qk5 3’ splice site compared to Gapdh mRNA (Fig 4D), indicating that Qk5 is required for efficient 3’ splice site usage. These results support a hierarchical model whereby inefficient production of initially small amounts of Qk5 promote increased expression of Qk5 mRNA through increased terminal 3’ splice site usage, leading to increased Qk5 protein, which then increasingly promotes Qk6 and Qk7 RNA expression.
Qk5 inhibits its own expression by 3’UTR binding
Previous studies of mammalian Qk5 (Larocque et al. 2002) or its homolog fly HowL (Nabel-Rosen et al. 2002) suggest a general role for Qk5 as a repressor of gene expression through binding to 3’UTRs. Qk protein binds the Qk5 3’UTR (Hafner et al. 2010; Van Nostrand et al. 2016) in a region that overlaps with one very high confidence predicted binding site ((Paz et al. 2014) Fig S4D). To test the hypothesis that Qk5 might negatively regulate its expression through binding its 3’UTR (independent of its positively acting role through splicing, Figs 4C and 4D), we cloned the entire Qk5 3’UTR into a renilla luciferase reporter construct and either depleted all Qk proteins (siQk) or Qk6 and Qk7 (siQk6/7, N.B. siQk5 targets the reporter and was not used). Depletion of all Qk protein forms, but not Qk6/7 alone (Fig S4E), leads to a 27% increase in renilla protein expression (Fig S4F; p = 0.08). This suggests that Qk5 negatively regulates its own protein levels specifically, especially considering that this effect is not observed using the Qk6 3’UTR reporter (Fig S3E).
Close examination of the Qk5 3’UTR suggest several putative Qk regulatory binding motifs, but only one region contains three overlapping CLIP peaks (Fig S4D). To test the requirement of this motif for Qk5 3’UTR reporter abundance, we substituted TACTAAC (wild-type) with TGGTACC (mutant; see Fig S4D inset) in an otherwise unchanged renilla Qk5 3’UTR reporter construct and measured reporter protein abundances after transfection in C2C12 myoblasts. We find mutation of the Qk binding element UACUAAC results in an increase in reporter protein expression compared to wild type (Fig S4G; p = 0.07), similar to the observation upon depletion of Qk protein (Fig S4F). Together these results suggest that Qk5 binds the UACUAAC element in its own 3’UTR to negatively regulate its expression. Thus, Qk5 protein at low levels promotes additional Qk5 mRNA accumulation through splicing, but at high levels acts by binding its own 3’ UTR to limit expression through an unknown mechanism.
Qk6 negatively regulates Qk5 protein expression
If Qk5 protein promotes expression of both its own and Qk6 mRNA, then reciprocal influences may exist whereby Qk6 protein feeds back to control Qk5 levels. We performed quantitative western blotting using the Qk5-specific antibody, and detected a reproducible 26% increase in Qk5 protein levels when Qk6 is depleted (Fig 5A). This increase in Qk5 levels is accompanied by a significant increase in Qk5 activity as detected on alternative splicing events under Qk5 control (e. g. Rai14 exon 11 is more repressed, the Capzb exon 9 splicing reporter is activated (p < 0.05), and splicing efficiency of the terminal Qk5 3’ splice site increases ((p < 0.01) Figs 3B, 3C, 4C, and 4D (Hall et al. 2013)). This increase in Qk5 protein level and activity upon loss of Qk6 is accompanied by a consistent but modest reduction in Qk5 mRNA level relative to control (Fig 5B). This opposing relationship between RNA stability and translational efficiency has been previously described (Kawai et al. 2004), and suggests that Qk6 protein binding to the Qk5 mRNA stabilizes the transcript while simultaneously repressing its translation. The decrease in Qk5 RNA, together with the increase in Qk5 protein combine to produce a 37% increase in translational efficiency ((Vasudevan and Steitz 2007); p = 0.27) of Qk5 mRNA when Qk6 is depleted. These data indicate that Qk6 is a translational repressor of Qk5 mRNA, consistent with the function of Qk6 on other mRNAs in other cell types (Saccomanno L 1999; Zhao et al. 2010).
Qk6 positively regulates its own translation
During our initial overexpression experiments with Myc-Qk6 we noticed an increase in levels of endogenous Qk6 (Fig S1A). We repeated the overexpression of Myc-Qk6 in myoblasts and performed quantitative western blotting to measure both ectopic and endogenous Qk6. We observe a 2.5-fold increase in endogenous Qk6 protein (p < 0.05; Fig 5C), without a significant RNA level change, indicating a 2.43-fold increase in translational efficiency (p < 0.05; Fig 5C). No significant change is observed in endogenous Qk6 protein levels (Fig 5C) when Myc-Qk5 is overexpressed (Fig S5A), indicating that the ability of Qk5 to promote Qk6 RNA accumulation (Fig 3E) is saturated by endogenous amounts of Qk5 and cannot be further increased. We conclude that Qk6 promotes translation of its own mRNA, even as it represses translation of Qk5 mRNA (Fig 5A).
To exclude the hypothesis that ectopic Myc-Qk6 increases the level of endogenous Qk6 protein solely through a posttranslational mechanism such as increased homo-dimerization, we measured luciferase expression from the luciferase-Qk6 3’UTR reporter with and without Myc-Qk6 overexpression (Fig S5B). Consistent with the result described above for endogenous Qk6, overexpression of Myc-Qk6 leads to an increase in renilla protein level compared to control (Fig 5D; p = 0.1), suggesting Qk6 mediates this effect in part through direct translational control. Thus, the Qk6 3’UTR contains an autoregulatory element through which Qk6 activates its own translation. Taken together with the data in Figs 3 and 4, these results identify a cross-isoform and autoregulatory network in myoblasts where Qk5 is the most abundant isoform and required for all Qk expression; Qk6 is less abundant but promotes its own translation while inhibiting Qk5 translation. These results expose a novel autoregulatory network that helps explain how the different Qk RBP family members are controlled to regulate the functional output from this gene in myoblasts.
Qk protein isoform abundance in different cells and tissues
Qk5 is the predominant Qk isoform present in C2C12 myoblasts (Fig S1A). In contrast, Qk6 and Qk7 are the predominant isoforms in adult mouse brain (Hardy et al. 1996). From the cross-isoform regulatory interactions we identified in C2C12 myoblasts, we hypothesize that in cell types with higher Qk6 and lower Qk5 protein: 1) the requirement of Qk5 for Qk6 RNA accumulation is lessened due to amplified positive autoregulation by Qk6, and 2) Qk6 more efficiently represses the translation of Qk5 mRNA. To test this, we used quantitative western blotting (Fig S6A) to identify cells in which Qk5 protein abundance is lower and Qk6 protein abundance is higher relative to total Qk protein than observed in C2C12 myoblasts. C2C12 cells have a high Qk5/Qk6 ratio (with Qk7 nearly undetectable; Fig S1A; Fig 6A), and, consistent with previous findings (Hardy et al. 1996), adult mouse optic nerve and cerebellum tissue have a much lower Qk5/Qk6 ratio (Fig 6A). In contrast, rat C6 glioma (Benda et al. 1971) and CG4 oligodendrocyte precursor cells (Louis et al. 1992) have intermediate Qk5/Qk6 ratios, with different amounts of Qk6 or Qk7 (Fig 6A). We chose rat C6 glioma cells as a counter example to mouse C2C12 cells to test whether the autoregulatory controls observed above operate in cells with low Qk5/Qk6 ratios.
Cross isoform Qk regulatory control is conserved in C6 glioma cells
We treated C6 glioma cells with siRNAs to deplete either the Qk5 or Qk6/7 isoforms (these target sequences conserved in mouse and rat). Although overall siRNA-mediated depletion appears less efficient in these cells as compared to C2C12 cells, the loss of Qk6/7 upon depletion of Qk5 is evident, indicating that the hierarchical requirement of Qk5 for expression of Qk6/7 RNA (Fig 6E) and protein (Fig 6D) is conserved. The magnitude of the reduction of Qk6 protein in C6 glioma cells is less than that observed in C2C12 myoblasts (down 42% in C6 cells, 67% in C2C12 cells, Fig 6D compare to Fig S3A), however. We also observe an increase in Qk5 expression under Qk6/7 depletion in C6 glioma cells, indicating that the feedback repression of Qk6 on Qk5 is conserved. However, apparent translational efficiency increases by 2.6-fold in C6 glioma cells (p < 0.01) compared to 37% in C2C12 myoblasts (p = 0.27) in the absence of Qk6/7. Therefore, due to the higher relative concentration of Qk6 in C6 glioma cells the observed translational repression is more efficient. Taken together, these findings reveal that the Qk autogenous regulatory network operates at various settings in different cells (Fig 7A), where the Qk5/Qk6/7 isoform ratio can vary in a stable way, using parallel sets of auto- and cross regulatory control. The existence of multiple different stable settings of the Qk5/Qk6/7 ratio suggests that the Qk regulatory network is responsive to the influence of other regulatory proteins and RNAs.
Discussion
We initiated this study to understand how RBP family isoforms that recognize the same RNA sequence in different functional contexts (splicing, RNA transport, translation, decay) are regulated within a single cell so that each of these processes is appropriately supplied with the correct amount of RBP. We chose Qk because all isoforms arise form a single gene, to avoid complexities that exist with other RBP families expressed from multiple genes. In the case of Qk, isoforms are assigned roles by compartment: nuclear functions (splicing, RNA biogenesis) are executed by Qk5, whereas cytoplasmic functions (translational control, mRNA decay) are executed by Qk6. Homeostasis and control of the levels of each isoform is established through a network of auto- and cross regulatory controls whereby Qk5 is required for its own and Qk6 RNA expression from the nucleus, and Qk6 activates its own and represses Qk5 mRNA at the translational level (Figs 7A and 7B). The strength of these controls is tunable such that cells express different ratios of Qk5 and Qk6 in a stable characteristic way.
A division of labor between Qk protein isoforms
We previously identified Qk as a regulator of muscle-specific alternative splicing through binding to intronic ACUAA sequence motifs (Hall et al. 2013). Since the Qk5 isoform contains a NLS (Wu et al. 1999), it seemed natural to expect Qk5 to be responsible for splicing. Indeed, Qk5 is both necessary (Figs 3B and 3C) and sufficient (Fig 2B) for splicing regulation, whereas Qk6 and Qk7 are dispensable (Figs 3B, 3C, and S3A). Although artificially overexpressed Myc-Qk6 can activate splicing (Fig 1F), its nuclear localization (Fig 1E) is likely mediated by heterodimerization with endogenous Qk5 (Pilotte et al. 2001). A mutant Qk protein containing all the common Qk sequences but no isoform-specific C-termini (QkBody) is distributed throughout both the nucleus and cytoplasm of C2C12 myoblasts and suffices for splicing, whereas Qk6 does not (Fig 2). A comparison of the widespread distribution of QkBody protein with the predominantly cytoplasmic distribution of Qk6 indicates that the 8 Qk6-specific C-terminal amino acids encode a cytoplasmic retention or nuclear export signal (Nakielny and Dreyfuss 1999; Cook and Conti 2010). Thus, Qk5 appears assigned to nuclear functions, while Qk6 is responsible for cytoplasmic functions, and these functions are enforced by localization elements that reside within the isoform specific amino acids. This suggests that the Qk protein elements required to interact with the splicing, translation, export and decay machineries lie within the protein sequences common to all Qk isoforms, and that additional dissection of Qk protein sequences will be necessary to map the sites required for function in each of those processes.
We (Fig 1A) and others (van der Veer et al. 2013; de Bruin et al. 2016b) also observe some Qk6 that appears nuclear in untransfected cells, possibly but not necessarily mediated by formation of Qk5-Qk6 heterodimers. While it is tempting to speculate that such heterodimers may contribute some novel function, their existence at high levels would undermine the ability of cells to satisfy ongoing demands for Qk activity for splicing (or translation) due to normal fluctuation of Qk5 levels in the nucleus (or Qk6 levels in the cytoplasm). The independent control of Qk in each compartment would be lost, because isoform identity is blurred in heterodimers. The need to limit heterodimer levels may be one reason that typically either Qk5 or Qk6 but not both are highly expressed in the same cell, a condition that disrupts the isoform division of labor (Figs 2-5 and 7).
Auto- and cross isoform regulation balances quantitative output from the Qk gene
To identify processes that require specific isoforms in the presence of others, we sought to deplete cells of each single isoform of Qk protein in turn, to confirm the distinct function(s) of each. This effort was made more challenging by the strong dependence of Qk6 expression on expression of Qk5 (Fig 3). In analyzing the effect of depleting each Qk isoform on the expression of RNA and protein for itself and the other Qk isoforms, we found that Qk5 is required for nuclear accumulation of both its own transcripts and those encoding Qk6/7 (Fig 4). We also provide evidence that Qk6 represses translation of Qk5 mRNA but activates translation of its own mRNA (Fig 5). These results allowed us to build a model of Qk autogenous and cross-isoform regulation for C2C12 cells (Fig 7A), that also applies in cell types with a lower Qk5:Qk6 ratio than C2C12 myoblasts (Fig 6). This control network is more similar to a rheostat than a bistable switch where either Qk5 or Qk6 expression dominates, because it supports stable intermediate isoform mixtures from the Qk gene in different cell types (Fig 7A). The structure of this network is such that the same compartmentalization that constrains the functional role of each isoform is also used to create the auto- and cross-regulatory controls. Qk5 acts on its own RNA and that of Qk6/7 through nuclear processes, whereas Qk6 acts on both its own mRNA and that of Qk5 through translation and decay in the cytoplasm. This network controls both the total amount of Qk protein by combining a strong dependence of Qk6 expression on Qk5 with negative feedback of Qk5 by both Qk5 (at the RNA level) and Qk6 (at the translational level). Autoregulation by RBPs is common and can be imposed at different RNA processing steps including splicing (Wollerton et al. 2004; Lareau et al. 2007; Ni et al. 2007), polyadenylation (Dai et al. 2012), mRNA stability/decay (Ayala et al. 2011), and translation (de Melo Neto et al. 1995; Wu and Bag 1998). Although cross-paralog regulation exists between RBP family members encoded by separate genes (Boutz PL 2007; Spellman et al. 2007; Wang et al. 2012)), the Qk regulatory network highlights the role of compartmentalization in both functional specification and role in the regulatory network of RBP isoforms (Fig 7). Other RBP families may use similar mechanisms for autogenous and cross-isoform control but for those encoded by multiple genes, additional feedback on transcription are possible and may be necessary.
Since the quantitative output from the Qk network appears to be set at different Qk5:Qk6:Qk7 ratios in different cell types (Fig 6A), it must also be sensitive to regulation by other factors. For example, the RBP RBFOX2 reduces Qk7 expression by repressing alternative splicing in human embryonic stem cells (Yeo et al. 2009). Since RbFox2 is an abundant splicing regulator in C2C12 cells (Bland et al. 2010; Singh et al. 2014), this may explain why Qk7 is nearly undetectable (Figs S1A and 6A). This seems not to hold in mouse cerebellum, where both Qk7 (Fig 6A) and RbFox2 (Gehman et al. 2012) are abundant, therefore yet other cell-type specific factors are likely present. In another example, miR-214-3p targets regions common to both the Qk6 and Qk7 3’UTR (van Mil et al. 2012) that are not present in the Qk5 3’UTR, consequently downregulating Qk6 and Qk7 (Irie et al. 2016). There likely are additional means by which Qk isoform ratios can be regulated, but the Qk auto- and cross-regulatory isoform network uncovered here may represent the foundational structure on which that regulation is imposed.
Qk isoform ratio regulation, development, and cancer
Qk is a key regulator of development (Ebersole et al. 1996; Li et al. 2003; Justice and Hirschi 2010) and cancer (Novikov et al. 2011; Chen et al. 2012; Zong et al. 2014). Our model predicts that Qk5 expression precedes expression of other Qk isoforms temporally as transcription is upregulated at the Qk gene (Fig 7B), either during cancer progression or in a developmental context. This expectation appears to hold for embryogenesis in Xenopus (Zorn et al. 1997), zebrafish (Radomska et al. 2016), and mouse (Ebersole et al. 1996; Hardy et al. 1996). Additionally, early stem/progenitor cells in Drosophila (Nabel-Rosen et al. 1999) and mice (Hardy 1998), express predominantly Qk5 (or its ortholog HowL in Drosophila). In mice, qk knockout is embryonic lethal (Li et al. 2003) and the qkl1 mutant mouse, which appears to lack Qk5 expression due to disruption of Qk5-specific splicing by a point mutation (Cox et al. 1999), is recessive lethal (Shedlovsky et al. 1988). In this mutant, both Qk6 and Qk7 mRNAs are significantly reduced in visceral endoderm dissected from non-viable embryos (Cox et al. 1999; Bohnsack et al. 2006), suggesting that Qk5 is required for efficient Qk6 and Qk7 expression in this tissue also. These observations provide additional evidence in support of our model of Qk autoregulation, and suggest its importance in an evolutionarily conserved developmental context.
The extensively studied qkv mouse model of dysmyelination (Sidman et al. 1964) expresses reduced mRNA levels of all Qk isoforms in glia (Lu et al. 2003) and vascular smooth muscle (van der Veer et al. 2013), due to a deletion that includes a portion of the Qk promoter (Ebersole et al. 1996). Based on the models we present here (Fig 7), lower Qk5 protein levels in qkv oligodendrocyte precursor cells may be insufficient for robust Qk6 and Qk7 up-regulation that normally occurs around the peak of myelination (Ebersole et al. 1996). Since deletion of qk in oligodendrocytes is lethal (Darbelli et al. 2016), the reduction in Qk6 and Qk7 in qkv oligodendrocytes could be due to selective pressure imposed on Qk5 substrates to maintain viability of the organism. Therefore, the imbalance of Qk isoforms in qkv oligodendrocytes results in developmental defects and the tremor phenotype without sacrificing viability of the organism.
Finally, disruption of tissue-specific Qk isoform expression patterns is also observed in non-small-cell lung cancer (NSCLC) (Sebestyen et al. 2015; de Miguel et al. 2016) and glioblastoma (Xi et al. 2017) patient samples. Although overall Qk is reduced, an isoform ratio switch occurs whereby healthy lung cells expressing mostly Qk6 transition to a state in NSCLC samples where cells express predominantly Qk5 (Sebestyen et al. 2015; de Miguel et al. 2016). This isoform switch likely shifts QKI function away from cytoplasmic regulation such as translation and decay (Saccomanno L 1999; Zhao et al. 2010; Figs 5 and 6), toward nuclear processes such as splicing (Figs 2B, 3B, 3C, 4C, and 4D) and primary transcript stability (Fig 4B). Understanding the relationship of QKI gene function in the context of cancer and other diseases requires evaluation of both the specific functions of Qk protein isoforms on their own RNAs as well as on the other pre-mRNAs and mRNAs from the many other genes that QKI regulates.
Materials and Methods
Cell Culture, Lysis, and Transfections
C2C12 cells were routinely cultured in Dulbecco’s modified Eagle medium (DMEM) with high glucose (Life Technologies) supplemented with 10% heat inactivated fetal bovine serum (Life Technologies) at 37°C with 5% CO2. For differentiation experiments, C2C12 cells were allowed to reach ∼90% confluency, then the media was changed to DMEM supplemented with 5% horse serum (Life Technologies) and the cells were harvested 72h post media change. C6 glioma cells were cultured in F-12K media (ATCC) supplemented with fetal bovine serum at 2.5% and horse serum at 15% as described (Benda et al. 1968) at 37°C with 5% CO2. CG4 oligodendrocyte precursor cells were grown in DMEM with high glucose supplemented with N1 (5 μg/ml transferrin, 100 μM putrescine, 20 nM progesterone, and 20 nM selenium), biotin (10 ng/ml), insulin (5 μg/ml), and 30% B104 cell conditioned media as described (Louis et al. 1992) at 37°C with 5% CO2; all supplements purchased from Sigma-Aldrich. Protein lysates were prepared by directly lysing cells in a buffer consisting of 100 mM Tris-HCl pH 7.4, 0.5% NP-40, 0.5% Triton X-100, 0.1% SDS, 100 mM NaCl, and 2.5 mM MgCl2. Cell lysis and analysis of luciferase protein expression for 3’UTR reporter assays were performed using the Dual Glo Luciferase Assay system (Promega) according to the manufacturer’s instructions (luciferase reporter RNA could not be accurately measured due to persistent contaminating plasmid DNA resistant to DNase treatment). RNA extraction was performed using TriZOL reagent (Life Technologies) according to the manufacturer’s suggestions and genomic DNA was degraded using Turbo DNase (Life Technologies) according to the manufacturer’s specifications. Transfections were performed using the Lipofectamine2000 reagent (Life Technologies) according to the manufacturer’s specifications.
Additional methods and oligonucleotide sequences found in Supplemental Information.
Figure S1 (supplemental to Figure 1): A. Representative western blot of whole cell extracts prepared from C2C12 myoblasts transfected as described in Fig 1E and 1F probed simultaneously with PanQk antibody (top; notation to the right indicates observed protein migration pattern) and Gapdh (bottom). B. Representative western blot of whole cell extracts prepared from C2C12 myoblasts transfected with control siRNA (siNT), or siRNA targeting all Qk isoforms (siQk), siQk plus Myc-Qk5, and siQk plus Myc-Qk6 Myc epitope antibody (top), Qk5 antibody (second from top), Qk6 antibody (second from bottom), and Gapdh antibody (bottom).
Figure S2 (supplemental to Figure 2): A. Representative western blot of protein lysates from C2C12 myoblast transfections described in Fig 2E and 2F: top shows blot probed with PanQk antibody (observed migration pattern of endogenous and mutant proteins shown at right), middle probed with HA epitope antibody, and bottom probed with Gapdh antibody.
Figure S3 (supplemental to Figure 3): A. Western blots simultaneously probed with Qk6 and Gapdh (top) or Qk7 and Gapdh (bottom) antibodies from whole cell protein lysate from C2C12 myoblasts transfected with siNT, siQk5-1, and siQk6/7; mean percentage of protein abundance relative to siNT is shown below from three independent biological replicates, +/− standard deviation from the mean. B. Representative western blot simultaneously probed with Qk5 and Gapdh (top) or Qk6 and Gapdh (bottom) antibodies from whole cell protein extracted from C2C12 myoblasts transfected with siNT, siQk5-1, siQk5-2, and siQk5-3 (as in Fig 3E). C. UCSC Genome Browser screen shot showing genomic region including Qk5 pre-mRNA (top), a portion of the Qk7 3’UTR (middle), and Qk6 3’UTR (bottom), with RBP Map predicted Qk binding sites, peaks from PanQKI eCLIP experiments performed in HepG2 and K562 cells, a peak from tagged QKI PAR-CLIP experiment performed in HEK293T cells, and mammalian conservation track. D. Schematic of Qk6 3’UTR dual luciferase assay (left) and representative western blot (right) from three independent replicate experiments using whole cell extracts from C2C12 myoblasts transfected with renilla luciferase-Qk6-3’UTR reporter and firefly luciferase with either siNT, siQk, or siQk5-1 probed with PanQk and Gapdh antibodies. E. Whole cell lysate from C2C12 myoblasts transfected as described in S3D analyzed for luciferase abundance: renilla luciferase normalized to firefly luciferase and expressed as fold change relative to siNT control; error bars represent the standard deviation from the mean of three independent biological replicates.
Figure S4 (supplemental to Figure 4): A. Schematic showing Qk5 RNA 3’end and primer sets used to measure total Qk5, unspliced Qk5, and read-through (non-polyadenylated) Qk5 measured in Fig 4C. B. Overview schematic of Qk5 3’ intron and terminal coding sequence splicing reporter (pDup-Qk5i) experimental strategy (used in Fig 4D); top: UCSC Genome Browser screen showing (from top to bottom) RefSeq gene annotation, RBP Map, PanQKI CLIP peaks in HepG2 cells, PanQKI CLIP peaks in K562 cells, and conservation tracks. Dotted lines denote region cloned into pDup51 splicing reporter to make pDup-Qk5i. C. Representative western blot of protein extracted from C2C12 myoblasts co-transfected with pDup-Qk5i and either siNT or siQk5-3, probed with PanQk antibody and Gapdh antibody. D. UCSC Genome Browser screen shot of Qk5 3’ end showing (from top to bottom) tracks for RefSeq, RBP Map, PanQKI CLIP peaks in HepG2 cells, PanQKI CLIP peaks in K562 cells, and QKI PAR-CLIP peak; inset below shows DNA encoding putative Qk binding motif ACTAAC (boxed, top, “wild type”) and the substitution mutant sequence (bottom, “mutant” used in S4E). E. Representative western blots using whole cell extract from C2C12 myoblasts transfected with wild type renilla luciferase Qk5 3’UTR reporter and firefly luciferase control and either siNT, siQk, or siQk6/7 probed with PanQk and Gapdh antibodies. F. Measurement of renilla luciferase containing full length wild-type Qk5 3’UTR co-transfected with control firefly luciferase containing bovine growth hormone 3’UTR and either control siNT, siQk, or siQk6/7; renilla luciferase protein was normalized to firefly luciferase protein and fold change relative to control values are reported from three independent biological replicate experiments; error bars represent standard deviation from the mean. G. Measurement of renilla luciferase Qk5 3’UTR reporter protein normalized to transfection control firefly luciferase protein from whole cell extracts from C2C12 myoblasts transfected in biological triplicate with either full length wild type or full length mutant Qk5 3’UTR reporter; data displayed as fold change relative to wild-type Qk5 3’UTR reporter and error bars represent standard deviation from the mean.
Figure S5 (supplemental to Fig 5): A. Representative western blot of C2C12 myoblasts mock transfected, transfected with Myc-Qk5, or Myc-Qk6 simultaneously probed with Qk5 (top) and Gapdh (bottom) antibodies. B. Western blots of whole cell protein lysates from C2C12 myoblasts co-transfected with renilla luciferase-Qk6 3’UTR reporter and firefly luciferase transfection control reporter plus either tdTomato control plasmid or Myc-Qk6 plasmid simultaneously probed with antibodies to DsRed2 (reacts with tdTomato), Qk6 and Gapdh.
Figure S6 (Supplementary to Fig 6): A. Representative western blots of whole cell protein extracts from C2C12 myoblasts, rat C6 glioma cells, rat CG4 oligodendrocyte precursor cells, mouse optic nerve tissue, and mouse cerebellum tissue probed with Qk5 in red channel and PanQk in green channel (top), Qk6 in red channel and PanQk in green channel (middle), and Qk7 in red channel and PanQk in green channel (bottom); signal intensities used to calculate data in Fig 6A. B. Representative western blot using whole cell protein extracts from rat C6 glioma cells transfected with siNT, siQk5-1, and siQk6/7 simultaneously probed with antibodies to Qk7 and Gapdh; Qk7 signal intensity normalized to Gapdh and reported as mean percentage protein abundance relative to siNT +/− standard deviation of the mean from 3 independent biological replicates.
Acknowledgements
We thank Sean Ryder, Doug Black, Jeremy Sanford, Rhonda Perriman, David Feldheim, and Susan Strome for reagents, discussion, and advice. This work was supported by National Institutes of Health (NIH) grant R01-GM040478 to M.A., NIH pre-doctoral T32 GM008646 to WSF, and California Institute of Regenerative Medicine (CIRM) pre-doctoral fellowship award TG2-01157 to WSF.