Skip to main content
NCBI home page
eLife logoLink to eLife
. 2016 Mar 8;5:e12876. doi: 10.7554/eLife.12876

Degradation of Gadd45 mRNA by nonsense-mediated decay is essential for viability

Jonathan O Nelson 1, Kristin A Moore 2,3, Alex Chapin 1, Julie Hollien 2,3, Mark M Metzstein 1,*
Editor: Torben Heick Jensen4
PMCID: PMC4848089  PMID: 26952209

Abstract

The nonsense-mediated mRNA decay (NMD) pathway functions to degrade both abnormal and wild-type mRNAs. NMD is essential for viability in most organisms, but the molecular basis for this requirement is unknown. Here we show that a single, conserved NMD target, the mRNA coding for the stress response factor growth arrest and DNA-damage inducible 45 (GADD45) can account for lethality in Drosophila lacking core NMD genes. Moreover, depletion of Gadd45 in mammalian cells rescues the cell survival defects associated with NMD knockdown. Our findings demonstrate that degradation of Gadd45 mRNA is the essential NMD function and, surprisingly, that the surveillance of abnormal mRNAs by this pathway is not necessarily required for viability.

DOI: http://dx.doi.org/10.7554/eLife.12876.001

Research Organism: <i>D. melanogaster</i>

eLife digest

Messenger RNA (mRNA) molecules act as the templates from which proteins are made, and so control the amount of protein in a cell. Having too much of certain proteins can harm cells. Additionally, some mRNAs contain errors, and so can create faulty proteins that may also harm the cell.

Cells have therefore developed ways to destroy excess or error-ridden mRNAs to avoid a deadly build up of proteins. One such quality control mechanism is called nonsense-mediated decay (NMD). This mechanism is so important that cells that cannot perform nonsense-mediated decay die, although it is not clear exactly what kills the cells.

Now, Nelson et al. have found that fruit flies whose cells are unable to perform nonsense-mediated decay die because a harmful protein called Gadd45 builds up in the cells. In normal cells, nonsense-mediated decay destroys the mRNA that relays the instructions for making Gadd45, which keeps the amount of the Gadd45 protein in the cell low. Further experiments show that removing Gadd45 from cells that lack nonsense-mediated decay saves the flies. Removing Gadd45 from human and mouse cells that are unable to perform nonsense-mediated decay also allows these cells to survive.

These findings imply that the only nonsense-mediated decay function needed for cells to live is the destruction of Gadd45 mRNA. This further implies that most faulty and normal mRNAs that are normally destroyed by nonsense-mediated decay do not cause the cells to die when nonsense-mediated decay is lost.

Learning that creating faulty proteins when nonsense-mediated decay is lost is not necessarily harmful to cells opens new possibilities to treating numerous genetic diseases. In some diseases, cells can only produce faulty forms of a particular protein. Nonsense-mediated decay normally destroys all of these mutant proteins, but it may sometimes be better to have faulty versions of a protein than to have none of it. Safely getting rid of nonsense-mediated decay by also eliminating Gadd45 from cells may therefore be a treatment strategy worth exploring.

DOI: http://dx.doi.org/10.7554/eLife.12876.002

Introduction

Maintaining proper gene expression is critical for normal development and physiology. In addition to de novo transcription, mRNA stability substantially contributes to forming the landscape of expression in a cell. The nonsense-mediated mRNA decay (NMD) pathway is a trans-acting mechanism that destabilizes mRNAs, and is best known for its well-described role as a quality control system, degrading abnormal mRNAs containing premature termination codons (PTCs) (Celik et al., 2015). NMD also degrades many wild-type endogenous mRNAs and thus is an important aspect of their post-transcriptional (Peccarelli and Kebaara, 2014). Loss of either of the core NMD genes Upf1 (Rent1) or Upf2 causes lethality in most eukaryotes (Kerényi et al., 2008; Medghalchi et al., 2001; Metzstein and Krasnow, 2006; Weischenfeldt et al., 2008; Wittkopp et al., 2009), indicating regulation of mRNA stability by NMD is critical for viability. However, the relative contributions to lethality from ectopic stabilization of PTC-containing mRNAs or endogenous NMD targets in NMD mutants remains unclear (Hwang and Maquat, 2011).

To identify which ectopically stabilized mRNAs are responsible for inducing lethality in NMD mutants, we performed an unbiased genetic suppressor screen seeking to restore viability in a Drosophila NMD mutant. To detect subtle increases in survival, we screened to suppress the lethality of animals mutant for the partially viable, hypomorphic Upf225G allele, of which 10% survive to adulthood (Chapin et al., 2014; Metzstein and Krasnow, 2006). We crossed this allele to heterozygous deficiencies to simultaneously reduce the mRNA abundance of several loci (Figure 1A). Of the 376 deficiencies tested, covering more than half the genome, ~10% suppressed NMD mutant lethality (Figure 1B, Figure 1—figure supplement 1A). The suppression effect could not be explained by a reduction in overall mRNA load, as there was only a weak correlation between the increase in mRNAs expressed from a genomic region upon loss of NMD function and the strength of suppression when that region was removed by a deficiency (Figure 1—figure supplement 1B). Rather, deficiencies that suppressed NMD-mutant lethality clustered in three genomic regions (Figure 1—figure supplement 1A). These findings suggest that NMD mutant lethality is not the result of a global excess of nonspecific mRNAs, but rather is mediated by specific genes residing within the few identified regions.

Figure 1. Drosophila suppressor screen identifies the Gadd45 pathway as the inducer of NMD-mutant lethality.

(A) Scheme to screen deficiencies for the suppression of Upf225G partial lethality. The Deficiency Suppression Score (DSS) represents the relative difference in Upf225G viability when crossed to a heterozygous deficiency (Df) compared to when crossed to a balancer (Bal) (See Methods). (B) DSS from 376 screened deficiencies ranked by score. A DSS greater than 0.1 (dotted line) indicates that deficiency suppresses Upf225G lethality. (C and E) Candidate suppressing regions uncovering Gadd45 (C) and Mekk1 (E). DSSs are shown in parenthesis. Dotted lines denote extent of regions deleted by suppressing deficiencies but not non-suppressing deficiencies. Filled blocks on chromosomes indicate predicted gene spans, Gadd45 pathway genes are indicated in red; suppressing deficiencies indicated in green, sple-J1 has undefined breakpoints located within hashed regions. (D and F) NMD mutant adult viability in combination with Gadd45F17 (D) or Mekk1Ur36 (F) mutants. Upf126A and Upf27-5A are null alleles (Frizzell et al., 2012; Metzstein and Krasnow, 2006). p-value compared to controls determined by the test of equal or given proportions indicated. Error bars represent 95% confidence interval of the binomial distribution. n equals total number of animals scored in each cross.

DOI: http://dx.doi.org/10.7554/eLife.12876.003

Figure 1.

Figure 1—figure supplement 1. Reduced expression of specific loci, not overall mRNA abundance, produces NMD mutant suppression by deficiencies.

Figure 1—figure supplement 1.

(A) Map of 376 autosomal DrosDel deficiencies with an isogenic background and molecularly defined breakpoints (Ryder et al., 2007) and eight other deficiencies used to further test candidate suppressing regions without overlapping DrosDel deficiencies. 39 total deficiencies suppress Upf225G lethality, shown in green. Regions deleted by any non-suppressing deficiencies were eliminated as candidate suppressing regions, removing false positives and reducing the size of the candidate intervals. The three candidate suppressing regions that are deleted only by suppressing deficiencies are indicated in black and labeled 1–3. (B) Each deficiency’s Deficiency Suppression Score (DSS) compared to percent increase in RNA abundance in Upf225G compared to wild-type from loci removed by that deficiency according to RNA-seq from Chapin et al. (2014). Trend line in red; statistics calculated using Pearson correlation test.
DOI: http://dx.doi.org/10.7554/eLife.12876.004

Figure 1—figure supplement 2. Drosophila Gadd45 is an endogenous direct NMD target.

Figure 1—figure supplement 2.

(A) Gadd45 mRNA expression in adults of the given genotypes measured by qRT-PCR. Gadd45 mRNA expression is increased 16.7-fold in Upf225G mutants, and is eliminated by Gadd45F17 mutants. p-values display one-sided Student’s t-test of indicated condition compared to control. Error bars represent 2 SEM. (B) Fluorescence of GFP transgenes with SV40, Act5C, or Gadd45 3’ UTRs expressed by UAS driven by Actin:GAL4 in Upf2+ or Upf225G third instar larvae. SV40 and Gadd45 3’ UTR constructs show significantly increased fluorescence in Upf225G animals compared to Upf2+, indicating NMD-dependent post-transcriptional degradation of mRNAs containing these UTRs. The Act5C 3’ UTR construct has similar fluorescence in both backgrounds, indicating NMD does not regulate the post-transcriptional stability of this UTR. Micrographs show dorsal views with anterior at top. (C) MAL-A2, traL, and Gadd45 5’ and 3’ fragment mRNA expression measured by qRT-PCR in pcm14 null mutants (Waldron et al., 2015) normalized to controls. Transcript structures of a non-NMD-target, maltase A2 (MAL-A2); a known NMD-targeted transcript, the non-sex specific isoform of transformer (traL) (Rehwinkel, 2005; Metzstein and Krasnow, 2006); and the Gadd45 transcript (note Gadd45 has no introns). Open boxes indicate UTRs; grey boxes indicate coding regions. NMD targeting initiates endonucleolytic cleavage near the stop codon (Gatfield and Izaurralde, 2004), producing 5’ and 3’ fragments with unprotected ends, which are then subjected to degradation by cytoplasmic 3’-to-5’ and 5’-to-3’ exonucleases, respectively. qRT-PCR primer pairs 5’ (red) and 3’ (blue) to the cleavage site can be used to differentially measure the quantity of these fragments. The Drosophila 5’-to-3’ exonuclease is encoded by the XRN1 homologue pacman (pcm), and fragments 3’ to an endonucleolytic NMD cleavage accumulate in Drosophila cells with reduced XRN1 activity (Gatfield and Izaurralde, 2004). The MAL-A2 3’ primers show no difference in relative expression in pcm14 mutants compared to the 5’ primers, while tra and Gadd45 have relatively increased levels of a 3’ fragment in pcm14 mutants, revealing endonucleolytic cleavage has occurred between the primer pairs, probably near the stop codon, indicative of NMD-initiated degradation. p-value between indicated samples using a two-sided Student’s t-test are displayed. ns indicates a p-value greater than 0.05. Error bars represent 2 SEM.
DOI: http://dx.doi.org/10.7554/eLife.12876.005

Figure 1—figure supplement 3. F17 is a null allele of Gadd45.

Figure 1—figure supplement 3.

(A) Gadd45F17 is an imprecise excision of the P-element P{EPg}HP20647, deleting a 894 bp region that includes the entire Gadd45 coding region. Gadd45E8 is a precise excision of the same P-element, leaving Gadd45 intact. Coding region in grey; untranslated regions in white. Arrowhead indicates direction of transcription. (B) Adult viability of control and Gadd45F17mutants. p = 0.4463 between Gadd45F17 and controls, using the test of equal or given proportions. Error bars represent 95% confidence interval of the binomial distribution. n equals total number of animals scored in each cross.
DOI: http://dx.doi.org/10.7554/eLife.12876.006

Figure 1—figure supplement 4. Loss of Gadd45 does not restore NMD activity in NMD mutants.

Figure 1—figure supplement 4.

(A) Expression of the endogenous NMD target transcript traL (Rehwinkel, 2005; Metzstein and Krasnow, 2006) in control male, Upf225G/Y, and Upf225G/Y; Gadd45F17animals, measured by qRT-PCR. There is no significant difference in traL expression between Upf225G/Y and Upf225G/Y; Gadd45F17animals. p-values determined by one-sided Student’s t-test between indicated conditions are displayed. ns indicates a p-value greater than 0.05. (B) Relative abundance of PTC-containing Adhn4 (Chia et al., 1987) and dHR783 (Fisk and Thummel, 1998) allele mRNAs compared to wild-type allele mRNA abundance in animals heterozygous for Adhn4or dHR783in each indicated genotype (stabilization of dHR783 was not determined in Upf225G/Y; Gadd45F17 animals). Neither reduction nor elimination of Gadd45 restored destabilization of these alleles. p-values determined by two-sided Student’s t-test between indicated conditions are displayed. ns indicates a p-value greater than 0.05. Error represents 2 SEM.
DOI: http://dx.doi.org/10.7554/eLife.12876.007
Open in a new tab

We expected that any specific genes mediating NMD-mutant lethality would have increased expression levels in an NMD mutant and be a direct NMD target. The only gene located within the suppressing regions to fit these criteria is Gadd45 (Figure 1C, Figure 1—figure supplement 2A–C) (Chapin et al., 2014). To determine if NMD targeting of Gadd45 mRNA is critical for viability, we generated a Gadd45 null allele, F17, which completely removes the Gadd45 coding region (Figure 1—figure supplement 3A) and eliminates Gadd45 mRNA expression (Figure 1—figure supplement 2A). As a heterozygote, Gadd45F17 suppressed Upf225Glethality as strongly as the corresponding deficiency identified by our screen (Figure 1D). We found that Gadd45F17 homozygous mutants are fully viable (Figure 1—figure supplement 3B), allowing us to test complete loss of Gadd45 for the suppression of NMD-mutant lethality. Homozygous Gadd45F17 restored full viability to Upf225Gmutants, and remarkably even partially suppressed the complete lethality observed in null Upf1 and Upf2 mutants (Frizzell et al., 2012; Metzstein and Krasnow, 2006) (Figure 1D). Importantly, neither reducing nor eliminating Gadd45 restored NMD function to Upf225G mutants, as measured by the expression of both an endogenous NMD target (Figure 1—figure supplement 4A) and PTC-containing mRNAs (Figure 1—figure supplement 4B).

In mammals, GADD45 activates the MTK1/MEKK4 kinase in a well-defined stress response pathway (Takekawa and Saito, 1998). Strikingly, the Drosophila MTK1 orthologue, Mekk1, resides within another Upf225G suppressing region (Figure 1E). Similar to Gadd45, we found that Mekk1 null mutants (Inoue et al., 2001) suppressed Upf1 and Upf2 mutant lethality (Figure 1F). This suppression was not as strong as that caused by a loss of Gadd45, revealing that although MEKK1 mediates NMD mutant lethality, it is likely that GADD45 has additional downstream effectors that influence viability. Overall, our findings reveal that increased Gadd45 mRNA stability is the major factor inducing NMD mutant lethality, primarily via increased MEKK1 activity.

Activation of MTK1 in mammals triggers a MAPK signaling cascade that promotes apoptosis (Takekawa and Saito, 1998). Over-expression of Gadd45 in Drosophila also induces apoptosis (Peretz et al., 2007). Interestingly, Drosophila cells lacking NMD function show excess cell death in a variety of tissues (Avery et al., 2011; Frizzell et al., 2012; Metzstein and Krasnow, 2006). To test if increased Gadd45 contributes to this excess death, we used TUNEL staining to examine cell death in wing imaginal discs from Upf225G mutant third instar larvae. This analysis revealed elevated levels of cell death compared to controls (Figure 2A, B, E), and this defect was completely suppressed by Gadd45F17 (Figure 2C–E). To confirm that, this effect was not specific to the Upf2 gene or 25G allele, we examined the wing discs in mutants of another essential NMD gene, Smg5. We found that Smg5 discs also showed elevated TUNEL signal, which was eliminated by loss of Gadd45 (Figure 2—figure supplement 1A–E). These results demonstrate that excess Gadd45 accounts for ectopic cell death in NMD mutant tissues.

Figure 2. Loss of Gadd45 suppresses NMD-mutant cell death.

(A to D) DAPI (blue) and (A’to D’) TUNEL (red) staining in late 3rd instar larval wing discs from control (A); Upf225G (B); Gadd45F17 (C); and Upf225G; Gadd45F17 (D) animals. (A’’ to D’’) are 4x view of outlined section at the base of the blade of the wing disc from A’-D’, respectively. Scale bar represents 100 μm. (E) Relative TUNEL signal in control and mutant wing discs, normalized to control. p-value between indicated samples using a two-sided Student’s t-test are displayed. ns indicates a p-value greater than 0.05. Error bars represent 2 SEM. n equals total number of discs scored. (F to I) w- eye clones in Gadd45+ and Gadd45F17 backgrounds. Dashed lines indicate clone boundaries. (J) Quantification of the fraction of the eye composed of w- cells in control and mutant eyes. p-values indicate differences between Gadd45 mutant and control in the same NMD background (indicated by horizontal bars) or NMD mutant and control in the same Gadd45 background (indicated by value above each individual bar), using a two-sided Student’s t-test. ns indicates a p-value greater than 0.05. Error bars represent 2 SEM. n = 20 eyes for all conditions.

DOI: http://dx.doi.org/10.7554/eLife.12876.008

Figure 2.

Figure 2—figure supplement 1. Loss of Gadd45 suppresses ectopic cell death in Smg5 mutant wing discs.

Figure 2—figure supplement 1.

(A to D) DAPI (blue) and (A’ to D’) TUNEL (red) staining in late 3rd instar larval wing discs from Smg5G115/+ (A); Smg5G115/C391 (B); Smg5G115/+ Gadd45F17 (C); and Smg5G115/C391 Gadd45F17 (D) animals. Smg5G115 and Smg5C391 are null Smg5 alleles (J.O.N. and M.M.M., unpublished). Scale bar represents 100 μm. (E) Relative TUNEL signal in wing discs, normalized to Smg5G115/+. p-values determined by two-sided Student’s t-test between indicated conditions are displayed. ns indicates a p-value greater than 0.05. Error bars represent 2 SEM.
DOI: http://dx.doi.org/10.7554/eLife.12876.009
Open in a new tab

To test if Gadd45-induced cell death is the only cellular defect in NMD mutants, we examined NMD function in the developing eye. NMD is required for proper development of eye cells, as clonal patches of NMD mutant cells in eyes are reduced in size (Frizzell et al., 2012; Metzstein and Krasnow, 2006). We found that Gadd45 is partially responsible for this defect, as the size of eye-cell clones lacking NMD activity in a Gadd45F17 background was increased, although not fully restored (Figure 2F–J). These results indicate that some, but not all, defects associated with loss of NMD are dependent on Gadd45.

Gadd45 is one of the few genes that is directly regulated by NMD in both flies and mammals (Huang et al., 2011; Tani et al., 2012; Viegas et al., 2007), raising the possibility that excess Gadd45 abundance may also contribute to the NMD-mutant lethality observed in mammalian cells (Azzalin and Lingner, 2006; Li et al., 2015; Medghalchi et al., 2001; Weischenfeldt et al., 2008). To test this hypothesis, we analyzed the effects of Gadd45 and Upf1 depletion in mouse NIH-3T3 cells. Gadd45b mRNA (also known as MyD118), which is expressed at least 10-fold higher than any other Gadd45 paralogue in these cells (Yue et al., 2014), was degraded rapidly in a partially Upf1-dependent manner after transcription was blocked with actinomycin D (Figure 3A), and had increased expression during Upf1 knockdown (Figure 3D), confirming it is sensitive to NMD. We found that transfection of 3T3 cells with siRNAs targeting Upf1 resulted in significant reduction in cell counts after 48 hr (Figure 3B), but co-transfection with siRNAs targeting both Upf1 and Gadd45b largely reversed this effect (Figure 3B). The reduction in cell counts was primarily due to increased cell death, as we found that ~25% of cells transfected with Upf1 siRNA were undergoing apoptosis (Figure 3C). Co-transfection of siRNA targeting Gadd45b almost entirely eliminated this increase (Figure 3C), indicating the excess apoptosis observed in Upf1-knockdown cells was mostly due to increased Gadd45 activity. However, while Gadd45b knockdown very greatly suppresses this excess death, it does not as fully rescue cell numbers, suggesting loss of NMD may lead to both Gadd45b-dependent cell death as well as a Gadd45b-independent effect on proliferation. This mirrors the conclusions we made about the partial suppression of cell number defects in the Drosophila eye. Importantly, Upf1 mRNA expression was equivalently reduced and the expression of the mammalian endogenous NMD targets Rassf1 and CRCP (Tani et al., 2012) was equivalently increased in both the single and double knockdown experiments (Figure 3D), indicating that the restoration of viability was not due to a recovery of NMD pathway activity.

Figure 3. Gadd45b mediates cell lethality in Upf1 siRNA knockdown 3T3 mouse embryonic fibroblasts.

(A) Relative Gadd45b mRNA expression measured by qRT-PCR in NIH-3T3 cells after 48 hr of control (black) or Upf1 (red) siRNA treatment and 0 to 2 hr of actinomycin D treatment, normalized to expression prior to actinomycin treatment. The half-life calculated for each decay curve is indicated. (B) Relative viable cell count of Upf1 and Gadd45b single and double siRNA treatment normalized to control siRNA. p-values display two-sided Student’s t-test between indicated conditions. (C) Quantification of apoptosis as measured by annexin V staining. p-values display two-sided Student’s t-test between indicated conditions. (D) Relative mRNA expression of Upf1, Gadd45b, and two mammalian endogenous NMD targets, Rassf1 and CRCP (Tani et al., 2012) measured by qRT-PCR in Gadd45b and Upf1 single and double siRNA knockdown cells, normalized to expression in the control siRNA condition. p-values display one-sided Student’s t-test for each condition compared to control. Error bars represent 2 SEM.

DOI: http://dx.doi.org/10.7554/eLife.12876.010

Figure 3.

Figure 3—figure supplement 1. GADD45A mediates cell lethality in Upf1 knockdown HEK293 cells.

Figure 3—figure supplement 1.

(A) Relative GADD45A mRNA expression measured by qRT-PCR in UPF1 and GADD45A single and double siRNA knockdown 72 hr after siRNA transfection in HEK293 cells, normalized to expression in the control siRNA condition. p-values display one-sided Student’s t-test for each condition compared to control. (B) Viable cell count of UPF1 and GADD45A single and double siRNA-treated cells normalized to control siRNA -treated cells. p-values display two-sided Student’s t-test between indicated conditions. (C) Relative UPF1 mRNA expression measured by qRT-PCR in UPF1 and GADD45A single and double siRNA knockdown, normalized to expression in control siRNA condition. p-values display one-sided Student’s t-test for each condition compared to control. Error bars represent 2 SEM.
DOI: http://dx.doi.org/10.7554/eLife.12876.011
Open in a new tab

To extend our analysis to other mammalian cells, we analyzed the role of Gadd45 mediating the effects of loss of NMD in HEK293 cells. We found, similarly to 3T3 cells, that siRNA knockdown of UPF1 in HEK293 cells led to increased GADD45A expression and reduced cell numbers compared to control siRNA (Figure 3—figure supplement 1A,B). Although transfection of siRNA targeting GADD45A alone slightly reduced HEK293 cell numbers, co-transfection with UPF1 siRNA did not further reduce cell count (Figure 3—figure supplement 1B), and UPF1 expression was equivalently reduced in the single and double knockdown conditions (Figure 3—figure supplement 1C). These results suggest that UPF1 knockdown is no longer detrimental to HEK293 cell viability in the absence of GADD45A expression. We conclude that increased expression of mammalian Gadd45 genes contributes to lethality in NMD-deficient mouse and human cells, as Gadd45 does in Drosophila.

Deconvoluting the contributions to organismal viability of the PTC-surveillance versus gene-regulatory functions of NMD has been historically difficult (Hwang and Maquat, 2011). Here, we show that viability can be restored to Drosophila lacking core NMD factors when a single endogenous NMD target, Gadd45, is eliminated, and that the requirement for the regulation of Gadd45 by NMD is evolutionarily conserved from flies to mammals. Although our data suggest that up-regulation of Gadd45 is a major factor contributing to lethality when NMD activity is lost, it is likely that other NMD targets also contribute to the observed lethality. In particular, viability is not restored to 100% in null Upf1; or Upf2; Gadd45 double mutants. In addition, loss of Gadd45 suppresses programmed cell death caused by defects in NMD, but not additional cell cycle defects, as implied by the incomplete suppression in the Drosophila eye and mammalian cell culture. Such defects in the cell cycle may be particularly pronounced during the development of certain tissue, or specific developmental stages. Indeed, NMD has been reported to have differing stage and tissue- specific activities (Bao et al., 2015; Bruno et al., 2011; Colak et al., 2013; Li et al., 2015). Whether this is due to a role in surveillance or another specific target remains unclear, but examination of the effects of loss of NMD in Gadd45 mutants should allow exploration of these possibilities.

The benefit for such a mechanism regulating Gadd45 expression may lie in a function of NMD in restricting viral growth (Balistreri et al., 2014). Because viruses encode trans-acting factors to inhibit NMD (Mocquet et al., 2012), the resulting accumulation of GADD45 in infected cells may act as a “molecular tripwire” that rapidly elicits a stress response and cell death. This outcome suggests that regulating responses to infection may underlie a conserved essential function of NMD. Intriguingly, restriction of pathogens via NMD extends to plants (Garcia et al., 2014), where NMD mutant lethality in A. thaliana, which do not encode Gadd45 orthologues, may be caused by the overexpression of a subset of immune-related intracellular nucleotide-binding leucine-rich repeat receptors, some of which are endogenous NMD targets (Gloggnitzer et al., 2014). In contrast, eukaryotes that do not rely on the activation of programmed cell death to protect against viruses, such as S. cerevisiae, S. pombe, and C. elegans, do not require NMD for viability (Hodgkin et al., 1989; Leeds et al., 1991; Mendell et al., 2000). Together these observations suggest a potential novel role for NMD and Gadd45 in immune responses, triggering the death of infected cells during pathogenic challenges.

Restoring the expression of PTC-containing alleles via NMD inhibition has been proposed as a promising therapy for a wide range of recessive genetic diseases (Keeling et al., 2014). Translation of stable PTC-containing mRNAs would produce truncated proteins that may be partially functional and alleviate disease symptoms normally caused by complete loss of the protein. However, the essential function for NMD in viability has raised the concern that these therapies may have prohibitive side effects. Our findings reveal a molecular basis for dealing with this obstacle by suggesting that inhibiting both the NMD and Gadd45 pathways (Tornatore et al., 2014) in combination could provide an effective and safe treatment for patients with debilitating genetic disorders.

Materials and methods

Fly genetics

Drosophila melanogaster stocks were raised on standard cornmeal/dextrose food at 25°. The NMD mutant alleles Upf225G, Upf27-5A, and Upf126A(Frizzell et al., 2012; Metzstein and Krasnow, 2006) are on y w FRT19A chromosomes. These alleles were balanced over FM7i, P{ActGFP}JMR3 (Reichhart and Ferrandon, 1998). Smg5G115 and Smg5C391 are null alleles of Smg5 (J.O.N., D. Förster, S. Luschnig, and M.M.M., unpublished) and will be described in detail later. The Smg5 alleles are balanced over CyO, P{Dfd:eYFP w+} (Le et al., 2006). Other alleles used were P{w[+mC]=EPg}HP20647 (Staudt et al., 2005), Mekk1Ur36(Inoue et al., 2001) recombined on FRT82Bby D. Ryoo, ey-FLP (Newsome et al., 2000), pcm14 (Waldron et al., 2015), Adhn4(Chia et al., 1987) and DHR783 (Fisk and Thummel, 1998). Control chromosomes were y w FRT19A (for Upf1 and Upf2) and FRT82B (for Mekk1) (Xu and Rubin, 1993). For all experiments using Gadd45F17 we used the Gadd45E8 precise excision as a control.

For viability assays, we mated flies for 3 days and collected all progeny each day for 10 days, starting 10 days after the cross was initiated. The total numbers of F1 mutant and balancer males were scored, and the ratio of mutant males to balancer males was used to determine mutant animal viability. To control for balancer viability within each experiment, we normalized the ratio of mutant to balancer animals to a ratio of the appropriate control chromosome to balancer animals produced from a parallel cross.

Deficiency suppressor screen

We screened autosomal deficiencies from the DrosDel collection (Ryder et al., 2007). All deficiencies scored can be found in Supplementary file 1. Deficiencies on chromosome 2 were balanced over CyO, and deficiencies on chromosome 3 were balanced over TM6C. We mated males from each deficiency stock to y w Upf225G FRT19A/FM7i, P{ActGFP}JMR3 females and scored all F1 males for the presence or absence of each balancer. For any given deficiency tested, the percentage of Deficiency / + males that are Upf225G mutants, less the percentage of Balancer / + males that are Upf225G mutants was calculated, producing a Deficiency Suppression Score (DSS), which represents the effect of an individual deficiency on the increase or decrease in Upf225G viability, while controlling for each deficiency’s general influence on viability. A DSS greater than 0.1 indicates suppression of lethality. Supplemental deficiencies used were from the Exelixis collection (Parks et al., 2004) and Df(2R)sple-J1 (Heitzler et al., 1993). Deficiency mapping to the Drosophila genome was performed using the 5.1 genome release.

RNA-seq data sets were acquired from Chapin et al. (2014) (archives SRR896609, SRR896616, SRR503415, and SRR503416) and aligned using Bowtie and TopHat alignment with standard remapping parameters to the 5.1 Drosophila genome release. SAMtools accessory scripts were used to retrieve read counts for deficiency and control regions. All read counts were normalized to reads per million within each data set. Average normalized reads in Upf225G samples were normalized to the relative reads of 74 ribosomal proteins in Upf225G samples compared to control samples. Total normalized reads within the regions removed by each deficiency were averaged between biological replicates, and the difference between the Upf225Gand control samples was divided by one million to determine percent increase in genomic load across each deficiency region.

Generation of Gadd45 mutants

We produced P-element excision lines from the P{w[+mC]=EPg}HP20647 P-element insertion line crossed to a Δ2–3 transposase stock. We mated F1 males containing the P-element and transposase on a CyO balancer to w; Tft / CyO females. Cy+ Tft white-eyed F2 males were then individually mated to w; Tft / CyO females. We then collected Tft+, Cy males and females to create an isogenic stock from each individually mated F2 male. To identify precise excisions we used the primers Gadd45_F1 / Gadd45_R1 flanking the P-element insert site to amplify a region across the excised P-element. Lines that failed to amplify with these primers were candidate imprecise excisions, which we then tested with Gadd45_F1 / Gadd45_R3 primers for deletions. Any detected deletions were subsequently sequenced using these same primers. Primer sequences are found in Supplementary file 2.

Induction and analysis of eye clones

We generated eye clones with the FLP/FRT system using the ey-FLP driver (Newsome et al., 2000) to induce recombination. We imaged eyes on a Leica MZ125 stereo microscope with a Retiga-2000R camera (QImaging, Canada) with QCapture 3.1.2 software (QImaging). We focused images using the ImageJ stack focuser plugin and quantified relative eye clone size using the ImageJ analyzer tools. A total of 20 eyes from 20 individual animals were scored for each condition.

Cell death assays

For TUNEL assays, third instar larval wing discs were dissected as described in Sullivan et al. (Sullivan et al., 2000). TUNEL staining was performed using the Apoptag Red in situ Apoptosis Detection Kit (Chimicon International Inc., Billerica, MA) according to Chakraborty et al. (Chakraborty et al., 2015). We DAPI stained wing discs (1:5000) for 5 min prior to mounting. Confocal images were acquired using a Zeiss LSM710 laser scanning confocal microscope (Carl Zeiss AG, Germany). 3-dimensional datasets were acquired with a Plan-Apochromat 20X/0.8 lens, 1.34 μm z-step, using the Zeiss ZEN software. To measure TUNEL signal intensity z-projections images were summed with ImageJ. Background signal was removed by using the ImageJ MaxEntropy auto-threshold. Relative total TUNEL signal intensity was calculated using the ImageJ analyzer tools to measure the total pixel intensity within the wing discs of TUNEL images and normalized to the average intensity in control conditions.

For annexin V staining, we collected media (including floating cells) from siRNA treated cells. We spun down media at 950g for 4 min to pellet cells, and then aspirated remaining media. Concurrently, we trypsinized siRNA-treated cells still on plates and added them to the same respective tube as previously spun-down media. Following the Alexa Fluor 488 Annexin V/Dead Cell Apoptosis Kit (Abcam, UK) protocol, we stained for apoptotic cells. We visualized cells on an Olympus IX51 microscope (Olympus, Japan) with 20X objective. We collected bright field as well as fluorescent images using a FITC filter with a QImaging QICam Fast1394 camera and QCaptureP software (QImaging). We analyzed cells by counting all cells within a bright field image as well as the annexin V positive cells from the same image. The number of annexin V positive cells was divided by total cell number to generate the fraction of apoptotic cells for each treatment. >3000 total cells were counted across three biological replicates for each treatment.

Cell culture experiments

We cultured mouse NIH-3T3 cells (ATCC) or HEK293 cells (ATCC) in DMEM (Thermo-Fisher, Waltham, MA) supplemented with 10% fetal bovine serum and glutamine. For siRNA experiments, we transfected cells using RNAiMax and 24 pmol of negative control siRNA (Qiagen, Netherlands), Upf1 siRNA (Qiagen), or Gadd45b siRNA (Sigma-Aldrich) for 3T3 cell experiments, or negative control siRNA (Qiagen), UPF1 siRNA (Qiagen), or GADD45A siRNA (Sigma-Aldrich, St. Louis, MO) for HEK293 experiments. For double siRNA-treated cells, we used 24 pmol of each Upf1 and Gadd45b siRNA for 3T3 experiments or UPF1 and GADD45A siRNA for HEK293 experiments.

For actinomycin experiments, we incubated cells with siRNA for 48 hr before changing the media and then incubated with 2 μg/mL actinomycin (Sigma-Aldrich) for 1 or 2 hr. mRNA half-life was determined by fitting an exponential decay curve to the relative expression at each time point (Tani et al., 2012). t1/2 was calculated based on the average expression at each time point, and the mean t1/2 for each condition is represented.

For cell counting experiments, we trypsinized cells, incubated a small aliquot with Trypan Blue at a final concentration of 0.04% in complete media, and counted Trypan Blue negative cells. RNA was collected from the remaining cells, and relative mRNA levels were measured as described below.

RNA isolation and quantitative RT-PCR

For Drosophila qRT-PCR analyses, we collected 5–10 adult animals frozen in liquid nitrogen. We isolated total RNA using TRIzol reagent (Invitrogen) and phase-lock tubes (5-Prime), and the RNeasy mini kit (Qiagen). We used on-column RNase-free DNase treatment (Qiagen) to reduce genomic contamination. We determined RNA concentration by spectrophotometer and normalized concentration for reverse transcription. For reverse transcription, we used random decamers and MMLV8 reverse transcriptase (Retroscript Kit, Thermo-FIsher). We performed qRT-PCR analysis using the SYBR Green qPCR Supermix (Bio-Rad, Hercules, CA) and the Bio-Rad iCycler thermocycler. All experimental reactions were performed using three technical replicates and a minimum of three biological replicates per condition, and the expression level of all experimental assays was normalized to RpL32 mRNA expression.

For cell culture qRT-PCR analyses, we collected RNA following the Zymo Research Quick RNA MiniPrep kits protocol, and synthesized cDNA using MMLV reverse transcriptase (NEB, Ipswich, MA) with a template of 1 µg of total RNA and priming with a T18 oligo. We measured relative mRNA levels by qRT-PCR using the Masterplex ep realplex (Eppendorf, Germany) with SYBR green fluorescent dye. Each sample was measured with technical triplicates and three biological replicates, and target mRNA levels were normalized to those of ribosomal protein 19 (Rpl19) mRNA.

For all qRT-PCR analyses we also measured samples that had been made without reverse transcriptase to ensure that signal was not due to genomic DNA. Primer sequences can be found in Supplementary file 2.

3’ UTR cloning and sensitivity assay

We cloned the UAS-GFP::Gadd45 3’ UTR and control UAS-GFP::Act5C 3’ UTR constructs using the primers G45_3U_X1_F / G45_3U_S1_R or Act5C_X1_F / Act5C_S1_R (Supplementary file 2) to amplify the Gadd45 and Act5C 3’ UTRs, respectively, from genomic DNA. PCR fragments were inserted into the Zero Blunt TOPO vector (Thermo-Fisher), sequenced to assure fidelity, and digested and cloned into a pUAST-attB GFP vector using standard cloning procedures to replace the SV40 3’ UTR. Plasmids were injected by BestGene (Chino Hills, CA) into a stock containing the VK00027 attP site (Venken et al., 2006) for phiC31 directed integration. We used previously described UAS-GFP::SV40 3’ UTR animals (Metzstein and Krasnow, 2006). For imaging, wandering late L3 larvae were collected and examined using a Leica MZ 16F microscope and the Leica DFC340 FX camera with the Leica Application Suite v3.3.0 software.

Analysis of dHR783 and Adhn4 PTC allele stability

We collected adult F1 Upf2+; Gadd45E8/+, Upf225G; Gadd45E8/+, and Upf225G; Gadd45F17/+males that were also heterozygous for either the dHR783 or Adhn4. The Adhn4allele is a PTC-containing allele and has been demonstrated to be a direct NMD target based on cleavage by Smg6 (Gatfield and Izaurralde, 2004). The dHR783 allele is also a PTC-containing allele and thus is presumably degraded by NMD (Fisk and Thummel, 1998). At least three biological replicates were collected for each condition. We isolated RNA and generated cDNA as described in methods above and used this cDNA as a template for PCR amplification of the dHR78 transcript with the DRH78_F3 / DHR78_R3 primers and the Adh transcript with the Adh_F and Adh_R primers (Supplementary file 2), which flank the nonsense mutation in the respective transcripts. To compare the relative abundance of the dHR783 allele to the wild-type allele, PCR products were Sanger sequenced, and the relative peak intensity for a T (dHR783 allele) compared to a C (wild-type allele) at nucleotide 1063 was compared. To compare the relative abundance of the Adhn4 allele to the wild-type allele, PCR products were digested with PvuII (a site disrupted by the n4 mutation), separated on a 1% agarose gel and stained with ethidium bromide. The relative intensity of the cut and uncut bands was determined using ImageJ and normalized for fragment length. All samples were ran on the same gel and compared under identical conditions. All ratios were normalized to the ratio in the Upf225G; Gadd45E8/+condition.

Statistical analysis

All figures displaying viability assays represent a proportion of animals of the indicated genotypes that survive to adulthood; error bars for these figures represent the 95% confidence interval of the binomial distribution, and the Test of qual or Given Proportions was used to determine significance difference in these proportions between genotypes. All other figures represent the mean value of multiple replicates have error bars depicting ± 2 SEM, which is a close approximation of the 95% confidence interval (Krzywinski and Altman, 2013). For tests between two variable measures, a two-sided paired Student’s t-test was used to determine significance difference between mean value data. For most qPCR experiments, data was compared to a normalized control, set to a constant of 1, so these tests were performed with a one-sided Student’s t-test.

Acknowledgements

We thank the Metzstein and Thummel labs for helpful discussion; Shawn Rynearson, Esther Ellison, Zev Kronenberg, and EJ Osborne for assistance with collection and analysis of the deficiency suppressor screen; Kate Sanders for assistance isolating the Gadd45F17 allele; Don Ryoo, Stefan Luschnig, Sarah Newbury, and Carl Thummel for providing Drosophila lines; Kim Frizzell for assistance imaging Drosophila eyes; Ria Chakraborty and Kent Golic for assistance with TUNEL; Chase Bryan for assistance with confocal microscopy; and Carl Thummel, Gillian Stanfield, and Nels Elde for helpful comments on the manuscript. Fly stocks were obtained from the Bloomington Drosophila Stock Center. Exelixis deficiencies were provided by Exelixis, Inc. This work was supported by National Institutes of Health (NIH) grant 1R01GM084011 (to MMM) and a March of Dimes Award 5-FY07-664 (to MMM).

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Funding Information

This paper was supported by the following grants:

  • National Institutes of Health 1R01GM084011 to Jonathan O Nelson, Mark M Metzstein.

  • March of Dimes Foundation 5-FY07-664 to Mark M Metzstein.

Additional information

Competing interests

The authors declare that no competing interests exist.

Author contributions

JON, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

KAM, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

AC, Conception and design, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

JH, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

MMM, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article.

Additional files

Supplementary file 1. Deficiencies used in deficiency suppressor screen.

DOI: http://dx.doi.org/10.7554/eLife.12876.012

elife-12876-supp1.xlsx (54KB, xlsx)
DOI: 10.7554/eLife.12876.012
Supplementary file 2. PCR primers used in this study.

DOI: http://dx.doi.org/10.7554/eLife.12876.013

elife-12876-supp2.xlsx (11.6KB, xlsx)
DOI: 10.7554/eLife.12876.013

Major datasets

The following previously published dataset was used:

2014,D.melanogaster Nonsense-Mediated mRNA Decay study,http://trace.ncbi.nlm.nih.gov/Traces/sra/?study=SRP025939,http://trace.ncbi.nlm.nih.gov/Traces/sra/

References

  1. Avery P, Vicente-Crespo M, Francis D, Nashchekina O, Alonso CR, Palacios IM. Drosophila Upf1 and Upf2 loss of function inhibits cell growth and causes animal death in a Upf3-independent manner. RNA. 2011;17:624–638. doi: 10.1261/rna.2404211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Azzalin CM, Lingner J. The Human RNA Surveillance Factor UPF1 Is Required for S Phase Progression and Genome Stability. Current Biology. 2006;16:433–439. doi: 10.1016/j.cub.2006.01.018. [DOI] [PubMed] [Google Scholar]
  3. Balistreri G, Horvath P, Schweingruber C, Zünd D, McInerney G, Merits A, Mühlemann O, Azzalin C, Helenius A. The Host Nonsense-Mediated mRNA Decay Pathway Restricts Mammalian RNA Virus Replication. Cell Host & Microbe. 2014;16:403–411. doi: 10.1016/j.chom.2014.08.007. [DOI] [PubMed] [Google Scholar]
  4. Bao J, Tang C, Yuan S, Porse BT, Yan W. UPF2, a nonsense-mediated mRNA decay factor, is required for prepubertal Sertoli cell development and male fertility by ensuring fidelity of the transcriptome. Development. 2015;142:352–362. doi: 10.1242/dev.115642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bruno IG, Karam R, Huang L, Bhardwaj A, Lou CH, Shum EY, Song H-W, Corbett MA, Gifford WD, Gecz J, Pfaff SL, Wilkinson MF. Identification of a MicroRNA that Activates Gene Expression by Repressing Nonsense-Mediated RNA Decay. Molecular Cell. 2011;42:500–510. doi: 10.1016/j.molcel.2011.04.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Celik A, Kervestin S, Jacobson A. NMD: At the crossroads between translation termination and ribosome recycling. Biochimie. 2015;114:2–9. doi: 10.1016/j.biochi.2014.10.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chakraborty R, Li Y, Zhou L, Golic KG. Corp Regulates P53 in Drosophila melanogaster via a Negative Feedback Loop. PLOS Genetics. 2015;11:e12876. doi: 10.1371/journal.pgen.1005400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chapin A, Hu H, Rynearson SG, Hollien J, Yandell M, Metzstein MM. In Vivo Determination of Direct Targets of the Nonsense-Mediated Decay Pathway in Drosophila. G3: Genes|Genomes|Genetics. 2014;4:485–496. doi: 10.1534/g3.113.009357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chia W, Savakis C, Karp R, Ashburner M. Adhn4 of Drosophila melanogaster is a nonsense mutation. Nucleic Acids Research. 1987;15:3931. doi: 10.1093/nar/15.9.3931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Colak D, Ji S-J, Porse BT, Jaffrey SR. Regulation of Axon Guidance by Compartmentalized Nonsense-Mediated mRNA Decay. Cell. 2013;153:1252–1265. doi: 10.1016/j.cell.2013.04.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fisk GJ, Thummel CS. The DHR78 Nuclear Receptor Is Required for Ecdysteroid Signaling during the Onset of Drosophila Metamorphosis. Cell. 1998;93:543–555. doi: 10.1016/S0092-8674(00)81184-8. [DOI] [PubMed] [Google Scholar]
  12. Frizzell KA, Rynearson SG, Metzstein MM. Drosophila mutants show NMD pathway activity is reduced, but not eliminated, in the absence of Smg6. RNA. 2012;18:1475–1486. doi: 10.1261/rna.032821.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Garcia D, Garcia S, Voinnet O. Nonsense-Mediated Decay Serves as a General Viral Restriction Mechanism in Plants. Cell Host & Microbe. 2014;16:391–402. doi: 10.1016/j.chom.2014.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gatfield D, Izaurralde E. Nonsense-mediated messenger RNA decay is initiated by endonucleolytic cleavage in Drosophila. Nature. 2004;429:575–578. doi: 10.1038/nature02559. [DOI] [PubMed] [Google Scholar]
  15. Gloggnitzer J, Akimcheva S, Srinivasan A, Kusenda B, Riehs N, Stampfl H, Bautor J, Dekrout B, Jonak C, Jiménez-Gómez JM, Parker JE, Riha K. Nonsense-Mediated mRNA Decay Modulates Immune Receptor Levels to Regulate Plant Antibacterial Defense. Cell Host & Microbe. 2014;16:376–390. doi: 10.1016/j.chom.2014.08.010. [DOI] [PubMed] [Google Scholar]
  16. Heitzler P, Coulson D, Saenz-Robles MT, Ashburner M, Roote J, Simpson P, Gubb D. Genetic and cytogenetic analysis of the 43A-E region containing the segment polarity gene costa and the cellular polarity genes prickle and spiny-legs in Drosophila melanogaster. Genetics. 1993;135:105–115. doi: 10.1093/genetics/135.1.105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hodgkin J, Papp A, Pulak R, Ambros V, Anderson P. A new kind of informational suppression in the nematode Caenorhabditis elegans. Genetics. 1989;123:301–313. doi: 10.1093/genetics/123.2.301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Huang L, Lou C-H, Chan W, Shum EY, Shao A, Stone E, Karam R, Song H-W, Wilkinson MF. RNA Homeostasis Governed by Cell Type-Specific and Branched Feedback Loops Acting on NMD. Molecular Cell. 2011;43:950–961. doi: 10.1016/j.molcel.2011.06.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hwang J, Maquat LE. Nonsense-mediated mRNA decay (NMD) in animal embryogenesis: to die or not to die, that is the question. Current Opinion in Genetics & Development. 2011;21:422–430. doi: 10.1016/j.gde.2011.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Inoue H, Tateno M, Fujimura-Kamada K, Takaesu G, Adachi-Yamada T, Ninomiya-Tsuji J, Irie K, Nishida Y, Matsumoto K. A Drosophila MAPKKK, D-MEKK1, mediates stress responses through activation of p38 MAPK. The EMBO Journal. 2001;20:5421–5430. doi: 10.1093/emboj/20.19.5421. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Keeling KM, Xue X, Gunn G, Bedwell DM. Therapeutics based on stop codon readthrough. Annual Review of Genomics and Human Genetics. 2014;15:371–394. doi: 10.1146/annurev-genom-091212-153527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kerényi Z, Mérai Z, Hiripi L, Benkovics A, Gyula P, Lacomme C, Barta E, Nagy F, Silhavy D. Inter-kingdom conservation of mechanism of nonsense-mediated mRNA decay. The EMBO Journal. 2008;27:1585–1595. doi: 10.1038/emboj.2008.88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Krzywinski M, Altman N. Points of Significance: Error bars. Nature Methods. 2013;10:921–922. doi: 10.1038/nmeth.2659. [DOI] [PubMed] [Google Scholar]
  24. Le T, Liang Z, Patel H, Yu MH, Sivasubramaniam G, Slovitt M, Tanentzapf G, Mohanty N, Paul SM, Wu VM, Beitel GJ. A New Family of Drosophila Balancer Chromosomes with a w- dfd-GMR Yellow Fluorescent Protein Marker. Genetics. 2006;174:2255–2257. doi: 10.1534/genetics.106.063461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leeds P, Peltz SW, Jacobson A, Culbertson MR. The product of the yeast UPF1 gene is required for rapid turnover of mRNAs containing a premature translational termination codon. Genes & Development. 1991;5:2303–2314. doi: 10.1101/gad.5.12a.2303. [DOI] [PubMed] [Google Scholar]
  26. Li T, Shi Y, Wang P, Guachalla LM, Sun B, Joerss T, Chen Y-S, Groth M, Krueger A, Platzer M, Yang Y-G, Rudolph KL, Wang Z-Q. Smg6/Est1 licenses embryonic stem cell differentiation via nonsense-mediated mRNA decay. The EMBO Journal. 2015;34:1630–1647. doi: 10.15252/embj.201489947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Medghalchi SM, Frischmeyer PA, Mendell JT, Kelly AG, Lawler AM, Dietz HC. Rent1, a trans-effector of nonsense-mediated mRNA decay, is essential for mammalian embryonic viability. Human Molecular Genetics. 2001;10:99–105. doi: 10.1093/hmg/10.2.99. [DOI] [PubMed] [Google Scholar]
  28. Mendell JT, Medghalchi SM, Lake RG, Noensie EN, Dietz HC. Novel Upf2p Orthologues Suggest a Functional Link between Translation Initiation and Nonsense Surveillance Complexes. Molecular and Cellular Biology. 2000;20:8944–8957. doi: 10.1128/MCB.20.23.8944-8957.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Metzstein MM, Krasnow MA. Functions of the nonsense-mediated mRNA decay pathway in drosophila development. PLoS Genetics. 2006;2:e12876. doi: 10.1371/journal.pgen.0020180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mocquet V, Neusiedler J, Rende F, Cluet D, Robin J-P, Terme J-M, Duc Dodon M, Wittmann J, Morris C, Le Hir H, Ciminale V, Jalinot P. The Human T-Lymphotropic Virus Type 1 Tax Protein Inhibits Nonsense-Mediated mRNA Decay by Interacting with INT6/EIF3E and UPF1. Journal of Virology. 2012;86:7530–7543. doi: 10.1128/JVI.07021-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Newsome TP, Asling B, Dickson BJ. Analysis of Drosophila photoreceptor axon guidance in eye-specific mosaics. Development. 2000;127:851–860. doi: 10.1242/dev.127.4.851. [DOI] [PubMed] [Google Scholar]
  32. Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, Huppert K, Tan LR, Winter CG, Bogart KP, Deal JE, Deal-Herr ME, Grant D, Marcinko M, Miyazaki WY, Robertson S, Shaw KJ, Tabios M, Vysotskaia V, Zhao L, Andrade RS, Edgar KA, Howie E, Killpack K, Milash B, Norton A, Thao D, Whittaker K, Winner MA, Friedman L, Margolis J, Singer MA, Kopczynski C, Curtis D, Kaufman TC, Plowman GD, Duyk G, Francis-Lang HL. Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome. Nature Genetics. 2004;36:288–292. doi: 10.1038/ng1312. [DOI] [PubMed] [Google Scholar]
  33. Peccarelli M, Kebaara BW. Regulation of Natural mRNAs by the Nonsense-Mediated mRNA Decay Pathway. Eukaryotic Cell. 2014;13:1126–1135. doi: 10.1128/EC.00090-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Peretz G, Bakhrat A, Abdu U. Expression of the drosophila melanogaster GADD45 homolog (CG11086) affects egg asymmetric development that is mediated by the c-Jun N-Terminal kinase pathway. Genetics. 2007;177:1691–1702. doi: 10.1534/genetics.107.079517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rehwinkel J, Letunic I, Raes J, Bork P, Izaurralde E. Nonsense-mediated mRNA decay factors act in concert to regulate common mRNA targets. RNA. 2005;11:1530–1544. doi: 10.1261/rna.2160905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reichhart JM, Ferrandon D. Green balancers. Drosophila Information Service. 1998;81:201–202. [Google Scholar]
  37. Ryder E, Ashburner M, Bautista-Llacer R, Drummond J, Webster J, Johnson G, Morley T, Chan YS, Blows F, Coulson D, Reuter G, Baisch H, Apelt C, Kauk A, Rudolph T, Kube M, Klimm M, Nickel C, Szidonya J, Maroy P, Pal M, Rasmuson-Lestander A, Ekstrom K, Stocker H, Hugentobler C, Hafen E, Gubb D, Pflugfelder G, Dorner C, Mechler B, Schenkel H, Marhold J, Serras F, Corominas M, Punset A, Roote J, Russell S. The drosDel deletion collection: A drosophila genomewide chromosomal deficiency resource. Genetics. 2007;177:615–629. doi: 10.1534/genetics.107.076216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Staudt N, Molitor A, Somogyi K, Mata J, Curado S, Eulenberg K, Meise M, Siegmund T, Häder T, Hilfiker A, Brönner G, Ephrussi A, Rørth P, Cohen SM, Fellert S, Chung H-R, Piepenburg O, Schäfer U, Jäckle H, Vorbrüggen G. Gain-of-function screen for genes that affect drosophila muscle pattern formation. PLoS Genetics. 2005;1:e12876. doi: 10.1371/journal.pgen.0010055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sullivan W, Ashburner M, Hawley RS. New York, NY: Cold Spring Harbor Laboratory Press; 2000. Drosophila protocols. [Google Scholar]
  40. Takekawa M, Saito H. A family of stress-inducible GADD45-like proteins mediate activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell. 1998;95:521–530. doi: 10.1016/S0092-8674(00)81619-0. [DOI] [PubMed] [Google Scholar]
  41. Tani H, Imamachi N, Salam KA, Mizutani R, Ijiri K, Irie T, Yada T, Suzuki Y, Akimitsu N. Identification of hundreds of novel UPF1 target transcripts by direct determination of whole transcriptome stability. RNA Biology. 2012;9:1370–1379. doi: 10.4161/rna.22360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tornatore L, Sandomenico A, Raimondo D, Low C, Rocci A, Tralau-Stewart C, Capece D, D’Andrea D, Bua M, Boyle E, van Duin M, Zoppoli P, Jaxa-Chamiec A, Thotakura AK, Dyson J, Walker BA, Leonardi A, Chambery A, Driessen C, Sonneveld P, Morgan G, Palumbo A, Tramontano A, Rahemtulla A, Ruvo M, Franzoso G. Cancer-selective targeting of the NF-κB survival pathway with GADD45β/MKK7 inhibitors. Cancer Cell. 2014;26:495–508. doi: 10.1016/j.ccr.2014.07.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Venken KJT, He Y, Hoskins RA, Bellen HJ. P[acman]: A BAC transgenic platform for targeted insertion of large DNA fragments in D. melanogaster. Science. 2006;314:1747–1751. doi: 10.1126/science.1134426. [DOI] [PubMed] [Google Scholar]
  44. Viegas MH, Gehring NH, Breit S, Hentze MW, Kulozik AE. The abundance of RNPS1, a protein component of the exon junction complex, can determine the variability in efficiency of the Nonsense Mediated Decay pathway. Nucleic Acids Research. 2007;35:4542–4551. doi: 10.1093/nar/gkm461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Waldron JA, Jones CI, Towler BP, Pashler AL, Grima DP, Hebbes S, Crossman SH, Zabolotskaya MV, Newbury SF. Xrn1/Pacman affects apoptosis and regulates expression of hid and reaper. Biology Open. 2015;4:649–660. doi: 10.1242/bio.201410199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Weischenfeldt J, Damgaard I, Bryder D, Theilgaard-Monch K, Thoren LA, Nielsen FC, Jacobsen SEW, Nerlov C, Porse BT. NMD is essential for hematopoietic stem and progenitor cells and for eliminating by-products of programmed DNA rearrangements. Genes & Development. 2008;22:1381–1396. doi: 10.1101/gad.468808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wittkopp N, Huntzinger E, Weiler C, Sauliere J, Schmidt S, Sonawane M, Izaurralde E. Nonsense-mediated mRNA decay effectors are essential for zebrafish embryonic development and survival. Molecular and Cellular Biology. 2009;29:3517–3528. doi: 10.1128/MCB.00177-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Xu T, Rubin GM. Analysis of genetic mosaics in developing and adult Drosophila tissues. Development. 1993;117:1223–1237. doi: 10.1242/dev.117.4.1223. [DOI] [PubMed] [Google Scholar]
  49. Yue F, Cheng Y, Breschi A, Vierstra J, Wu W, Ryba T, Sandstrom R, Ma Z, Davis C, Pope BD, Shen Y, Pervouchine DD, Djebali S, Thurman RE, Kaul R, Rynes E, Kirilusha A, Marinov GK, Williams BA, Trout D, Amrhein H, Fisher-Aylor K, Antoshechkin I, DeSalvo G, See LH, Fastuca M, Drenkow J, Zaleski C, Dobin A, Prieto P, Lagarde J, Bussotti G, Tanzer A, Denas O, Li K, Bender MA, Zhang M, Byron R, Groudine MT, McCleary D, Pham L, Ye Z, Kuan S, Edsall L, Wu YC, Rasmussen MD, Bansal MS, Kellis M, Keller CA, Morrissey CS, Mishra T, Jain D, Dogan N, Harris RS, Cayting P, Kawli T, Boyle AP, Euskirchen G, Kundaje A, Lin S, Lin Y, Jansen C, Malladi VS, Cline MS, Erickson DT, Kirkup VM, Learned K, Sloan CA, Rosenbloom KR, Lacerda de Sousa B, Beal K, Pignatelli M, Flicek P, Lian J, Kahveci T, Lee D, Kent WJ, Ramalho Santos M, Herrero J, Notredame C, Johnson A, Vong S, Lee K, Bates D, Neri F, Diegel M, Canfield T, Sabo PJ, Wilken MS, Reh TA, Giste E, Shafer A, Kutyavin T, Haugen E, Dunn D, Reynolds AP, Neph S, Humbert R, Hansen RS, De Bruijn M, Selleri L, Rudensky A, Josefowicz S, Samstein R, Eichler EE, Orkin SH, Levasseur D, Papayannopoulou T, Chang KH, Skoultchi A, Gosh S, Disteche C, Treuting P, Wang Y, Weiss MJ, Blobel GA, Cao X, Zhong S, Wang T, Good PJ, Lowdon RF, Adams LB, Zhou XQ, Pazin MJ, Feingold EA, Wold B, Taylor J, Mortazavi A, Weissman SM, Stamatoyannopoulos JA, Snyder MP, Guigo R, Gingeras TR, Gilbert DM, Hardison RC, Beer MA, Ren B, Mouse ENCODE Consortium A comparative encyclopedia of DNA elements in the mouse genome. Nature. 2014;515:355–364. doi: 10.1038/nature13992. [DOI] [PMC free article] [PubMed] [Google Scholar]
eLife. 2016 Mar 8;5:e12876. doi: 10.7554/eLife.12876.016

Decision letter

Editor: Torben Heick Jensen1

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your work entitled "Degradation of a single mRNA by nonsense-mediated decay is essential for viability" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, with James Manley as the Senior Editor.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

A critical question in the RNA field is the functional significance of the nonsense-mediated RNA decay (NMD) pathway. NMD was originally discovered on the basis of its ability to degrade aberrant mRNAs (i.e., its RNA surveillance role) but was subsequently found to degrade many normal mRNAs (i.e., its gene expression role). Which is more important? With this elegant study, Mark Metzstein and colleagues take a great leap towards answering this question.

The authors sought to identify abnormally stabilized mRNAs that mediate lethality in NMD-defective Drosophila melanogaster. To this end, a genetic suppressor screen was performed in a Drosophila NMD mutant strain (hypomorphic Upf225G allele) and three candidate suppressing regions were identified. Two of these encompass the Gadd45 and Mekk1 genes, which operate in the same stress-response pathway. The importance of these genes for the NMD-lethality phenotype was analyzed further, and it was demonstrated that depletion of Gadd45 in the Upf225G-mutant completely restored viability, whereas it partially rescued viability in NMD null flies. Depletion of Mekk1, which acts downstream of Gadd45 in the stress response, also increased viability, but to a lesser degree. Next, it was demonstrated that Gadd45 is a direct target of NMD and that increased levels of Gadd45 can explain the increased cells death observed in the Upf225G-mutant.

Finally, the authors provided evidence that Gadd45b mRNA is also a direct NMD target in mouse NIH-3T3 cells and that its increased expression can partially explain the decreased viability observed in UPF1-depleted cells.

These observations led the authors to conclude that the essentiality of NMD in Drosophila and mammals in essence resides in the down-regulation of a single gene (Gadd45), which is involved in induction of apoptosis. The presented results, coupled with the finding of others that viral infection suppresses NMD, made the authors propose an interesting model in which NMD provides defense against virus infection by killing virally infected cells.

Essential revisions:

Overall, this study represents an important advance for the RNA/NMD field. Publication in eLife is therefore suggested if the authors can substantiate their conclusions. In its present form, the main conclusion (including the title) appears to be going somewhat too far. Several of the authors' own observations point to additional contributions to lethality besides higher Gadd45 expression – except in the specific Upf225G-mutant.

1) It is important to prove that the observed effects are coupled to increased Gadd45 protein. The authors need to show that expression of the Gadd45 protein is increased in Upf225G and perhaps also other NMD-defective cells.

2) Figure 2. It is critical that the authors examine suppression of apoptosis by Gadd45 in Upf1-mutant flies (e.g., Upf126A), not just Upf2-mutant flies (as both were examined in Figure 1). Given that NMD factors can function in pathways other than NMD, it is essential to assess more than one NMD factor.

3) Conservation in mammals. The results in Figure 3 are intriguing, but require some bolstering.

A) The mechanism underlying the alterations in cell counts must be investigated. Is it the result of alterations in cell death, proliferation, or both? Is the Gadd45-dependent apoptosis phenotype conserved in mammals?

B) At least one more cell line should be used for at least some experiments (one cell line is not enough to be convincing).

C) Given that there are several Gadd45 genes in mammals, it is important that these also be investigated. While it is appreciated that Gadd45b is the most highly expressed Gadd45 gene in NIH-3T3 cells, this does not rule out the importance of the other Gadd45 genes. Perhaps try the triple knockdown of Gadd45A, B and G. GADD45A appears to be the highest expressed variant in HEK293 and HeLa cells, so one or both of these cell lines could be obvious additional cell lines to test.

D) Figure 3A. One time point is not sufficient for RNA half-life analysis.

E) Figure 3C. More NMD substrates besides Smg5 should be used for this analysis. Indeed, the effect on Smg5 is very modest (probably because of either insufficient NMD factor knockdown or poor transfection efficiency). A lack of an effect on some NMD substrates is ok given the heterogeneity of the NMD pathway, but at least one NMD substrate should have a greater effect than only 50%. A cell line with greater transfection efficiency than NIH-3T3 might be necessary to use. The GAS5 non-coding RNA could be a possible candidate to check since it appears to be highly responsive to NMD in both human and mouse cells.

4) Although it is very clear that Gadd45 stabilization is a major determinant for the lethality phenotype of the Upf225G-mutant Drosophila strain and the decreased viability phenotype in UPF1-depleted NIH-3T3 cells, it is also clear that increased levels of Gadd45 cannot fully explain the lethality of NMD null mutant flies. Therefore, there must be other transcripts at play. One possibility is that NMD is vital for different stages of development, as e.g. indicated by several NMD knockout mouse studies. The authors should discuss more thoroughly the possibility that different stages of development may be affected by abnormal high expression of other factors than Gadd45. Generally, the conclusions (incl. the title) should be somewhat toned down, because overexpression of Gadd45 cannot fully explain lethality of all NMD-mutants. In relation to this: does the third deleted region contain genes that are related to the Gadd45 pathway or are they unrelated? In case of the latter, this also points to other contributors to lethality.

eLife. 2016 Mar 8;5:e12876. doi: 10.7554/eLife.12876.017

Author response

1) It is important to prove that the observed effects are coupled to increased Gadd45 protein. The authors need to show that expression of the Gadd45 protein is increased in Upf225G and perhaps also other NMD-defective cells.

We could not directly address Gadd45 protein expression in Upf225G flies because no antibodies that suitably detect Drosophila Gadd45 protein exist. We attempted to detect Drosophila Gadd45 using a broadly reacting antisera directed against HumanGADD45β, however, we found that this antibody does not cross-react with Drosophila Gadd45 protein. In addition, we tried to use this antisera in our mammalian cell culture, and also could not detect expression of endogenous Gadd45b (in the literature developed antibodies have only been tested on cells over-expressing Gadd45). However, since our data indicates that the loss of Mekk1 shows similar effects to loss of Gadd45 and that Gadd45 is known to regulate Mekk1 by a protein-protein interaction strongly suggests that Gadd45 protein levels must be increased in NMD mutants, even in the absence of a direct assay of Gadd45 protein levels in Drosophila. Furthermore, we show that a GFP reporter with the Gadd45 3’ UTR has increased fluorescence in Upf225G mutants, also suggesting that transcripts with the Gadd45 3’UTR have increased protein levels in NMD mutants.

2) Figure 2. It is critical that the authors examine suppression of apoptosis by Gadd45 in Upf1-mutant flies (e.g., Upf126A), not just Upf2-mutant flies (as both were examined in Figure 1). Given that NMD factors can function in pathways other than NMD, it is essential to assess more than one NMD factor.

We agree with the reviewers that it is important to test if loss of Gadd45 suppresses cell death in other NMD mutants, such as Upf1 or Upf2 null alleles. However, we have previously found that such mutants die during early larval stages(Chapin et al. 2014) and we were not able to isolate discs from these animals to examine cell death. Instead, we turned to another critical NMD gene, Smg5. In work we are currently preparing for publication we show that Drosophila Smg5 is critical for NMD function and viability, similar to Upf1 and Upf2, but Smg5 mutants die at a later stage. We examined TUNEL signal in the wing discs of Smg5 null mutants and found it is increased compared to controls and that Gadd45 mutants suppressed this ectopic cell death, very similarly to what we found in Upf225G mutant discs. This result indicates that the cell death in Upf2 mutant wing discs is in fact a defect due to loss of NMD pathway function and not the loss of NMD-independent Upf2 function. We are presenting this data in Figure 2—figure supplement 1. As a technical note, when analyzing the Smg5 wing discs our reagents lead to higher background signal than the previous experiments, so we had to adapt our quantification procedure (as described in the Methods section). In the interest of consistency, we reanalyzed our Upf225G data using this new approach and obtained essentially identical results to before. We have updated Figure 2 to represent this new analysis.

3) Conservation in mammals. The results in Figure 3 are intriguing, but require some bolstering.

A) The mechanism underlying the alterations in cell counts must be investigated. Is it the result of alterations in cell death, proliferation, or both? Is the Gadd45-dependent apoptosis phenotype conserved in mammals?

We have examined the mechanism for cell count changes by staining cells for the apoptotic marker annexin V. We find that Upf1 knockdown cells show a large increase in apoptosis, and this increase is almost entirely suppressed by simultaneous knockdown of Gadd45b. However, the increase in cell death probably does not alone account for the decrease in cell numbers, suggesting an underlying proliferation defect also occurs in NMD knockdown. Since Gadd45b knockdown appears to totally eliminate the apoptosis defect in NMD knockdown cells, but does not fully rescue cell numbers, this further suggests Gadd45b overexpression may not be responsible for the proliferation defect. We come to similar conclusions from Drosophila eye analysis. We have included these data in Figure 3, and in the Discussion (this is also relevant to comment 4, below)

B) At least one more cell line should be used for at least some experiments (one cell line is not enough to be convincing).

We have repeated our experiments using human HEK293 cells and obtained very similar results. These are shown in Figure 3—figure supplement 1.

C) Given that there are several Gadd45 genes in mammals, it is important that these also be investigated. While it is appreciated that Gadd45b is the most highly expressed Gadd45 gene in NIH-3T3 cells, this does not rule out the importance of the other Gadd45 genes. Perhaps try the triple knockdown of Gadd45A, B and G. GADD45A appears to be the highest expressed variant in HEK293 and HeLa cells, so one or both of these cell lines could be obvious additional cell lines to test.

We have attempted multiple Gadd45 knockdowns in both 3T3 and HEK cells, but we have found these to have deleterious effects on cell survival on their own, greatly convoluting our analysis of suppression of NMD knockdown. In these experiments we did find, as predicted by the reviewers, that GADD45A seems to primarily mediate the NMD knockdown response in HEK cells, so it is indeed the case that the predominantly expressed Gadd45 that seems to mediate NMD knockdown effects. However, since we show that we can rescue NMD cells by knockdown of a single homolog in both cell types, we believe the conclusions we present in the manuscript are valid. Knockdown of multiple Gadd45 paralogs, if it were possible, would be expected to further improve NMD cell survival, strengthening our conclusion, but the negative result of no increased survival would not suggest an alternative model.

D) Figure 3A. One time point is not sufficient for RNA half-life analysis.

We have added an additional time point and used this to calculate RNA half-life directly. This is presented as a revised panel in Figure 3A.

E) Figure 3C. More NMD substrates besides Smg5 should be used for this analysis. Indeed, the effect on Smg5 is very modest (probably because of either insufficient NMD factor knockdown or poor transfection efficiency). A lack of an effect on some NMD substrates is ok given the heterogeneity of the NMD pathway, but at least one NMD substrate should have a greater effect than only 50%. A cell line with greater transfection efficiency than NIH-3T3 might be necessary to use. The GAS5 non-coding RNA could be a possible candidate to check since it appears to be highly responsive to NMD in both human and mouse cells.

We have replaced the weakly affected Smg5 with two other substrates whose RNA levels are more than two-fold affected by NMD knockdown. We would have liked to have tested GAS5 directly, but all our current samples are polyA selected, so we unfortunately do not have this RNA represented.

4) Although it is very clear that Gadd45 stabilization is a major determinant for the lethality phenotype of the Upf225G-mutant Drosophila strain and the decreased viability phenotype in UPF1-depleted NIH3T3 cells, it is also clear that increased levels of Gadd45 cannot fully explain the lethality of NMD null mutant flies. Therefore, there must be other transcripts at play. One possibility is that NMD is vital for different stages of development, as e.g. indicated by several NMD knockout mouse studies. The authors should discuss more thoroughly the possibility that different stages of development may be affected by abnormal high expression of other factors than Gadd45. Generally, the conclusions (incl. the title) should be somewhat toned down, because overexpression of Gadd45 cannot fully explain lethality of all NMD-mutants. In relation to this: does the third deleted region contain genes that are related to the Gadd45 pathway or are they unrelated? In case of the latter, this also points to other contributors to lethality.

The reviewers make valid points about the likelihood of additional factors contributing to lethality in NMD mutant animals. We have incorporated these suggestions into the Discussion, and modified our title to “Degradation of Gadd45 mRNA by nonsense-mediated decay is essential for viability”, which more explicitly refers to the necessity of Gadd45 regulation by NMD. In addition, our data suggests that loss of NMD causes defects in cell survival via GADD45 upregulation, but also defects in the cell cycle independent of Gadd45. These latter defects do not always lead to cell lethality, but may contribute to the lower viability observed in NMD null mutants even with Gadd45 deleted. These points have been added to the Discussion.

Preliminary analysis of the third region indicates that suppression is likely due to multiple loci, and it is unclear if these loci function on the Gadd45 pathway (like Mekk1) or are independently regulated by NMD. We are continuing to analyze genes in this region and will report on it in a future publication.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    Supplementary file 1. Deficiencies used in deficiency suppressor screen.

    DOI: http://dx.doi.org/10.7554/eLife.12876.012

    elife-12876-supp1.xlsx (54KB, xlsx)
    DOI: 10.7554/eLife.12876.012
    Supplementary file 2. PCR primers used in this study.

    DOI: http://dx.doi.org/10.7554/eLife.12876.013

    elife-12876-supp2.xlsx (11.6KB, xlsx)
    DOI: 10.7554/eLife.12876.013

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES