Transcriptional adaptation in Caenorhabditis elegans.

Transcriptional adaptation is a recently described phenomenon by which a mutation in one gene leads to the transcriptional modulation of related genes, termed adapting genes. At the molecular level, it has been proposed that the mutant mRNA, rather than the loss of protein function, activates this response. While several examples of transcriptional adaptation have been reported in zebrafish embryos and in mouse cell lines, it is not known whether this phenomenon is observed across metazoans. Here we report transcriptional adaptation in C. elegans, and find that this process requires factors involved in mutant mRNA decay, as in zebrafish and mouse. We further uncover a requirement for Argonaute proteins and Dicer, factors involved in small RNA maturation and transport into the nucleus. Altogether, these results provide evidence for transcriptional adaptation in C. elegans, a powerful model to further investigate underlying molecular mechanisms.

Introduction

Transcriptional adaptation is the ability of certain mutations in a gene to modulate the expression of related genes, referred to as adapting genes (El-Brolosy and Stainier, 2017; El-Brolosy et al., 2019; Ma et al., 2019). At the molecular level, the mutant mRNA, rather than the loss of protein function, is responsible for this transcriptional modulation (Rossi et al., 2015; El-Brolosy and Stainier, 2017; Sztal et al., 2018; El-Brolosy et al., 2019; Ma et al., 2019). According to one model (El-Brolosy et al., 2019), the mutant mRNA, via its degradation products, modulates the expression of adapting genes via transcriptional regulators including antisense RNAs and histone modifiers. According to another model (Ma et al., 2019), the premature termination codon (PTC) containing mutant mRNA interacts with a histone modifier complex leading to transcriptional upregulation of the adapting gene(s). Sequence similarity with the mutant mRNA determines which genes get upregulated during transcriptional adaptation (El-Brolosy et al., 2019). In some cases, the upregulated genes share functionality with the mutated gene leading to functional compensation. However, while transcriptional adaptation is often discussed in the context of genetic robustness (Rossi et al., 2015; El-Brolosy et al., 2019; Ma et al., 2019), it does not always lead to functional compensation (Rossi et al., 2015). In addition, transcriptome analyses suggest that even genes with limited sequence similarity with the mutant mRNA can be upregulated during transcriptional adaptation (El-Brolosy et al., 2019), although clearly more work is required to determine whether the upregulation of these genes is a direct or indirect effect of transcriptional adaptation.

Understanding the mechanisms of transcriptional adaptation will help us better comprehend why for a given gene some mutations cause disease while others do not (Castel et al., 2018). However, despite the importance and growing interest in many aspects of genetic compensation, transcriptional adaptation has currently only been investigated in vertebrates. Thus, it remains unclear whether this phenomenon is observed across metazoans. The evolutionary importance of related genes that have compensatory effects has also been discussed in non-vertebrate eukaryotes (Conant and Wagner, 2004; Plata and Vitkup, 2014). However, it is not known whether these examples of compensation are due to protein feedback loops or transcriptional adaptation.

Only a few factors are known to be involved in the transcriptional adaptation response thus far, and others, including some involved in RNA processing and transport, are likely required. In addition, it is not clear whether the mechanisms of transcriptional adaptation are common or whether each particular case occurs in a different manner, especially at the step leading to transcriptional modulation. Also, as different paralogs or related genes are expressed in distinct tissues and/or at different times (Laisney et al., 2010; Kryuchkova-Mostacci and Robinson-Rechavi, 2015; Radomska et al., 2016; Pasquier et al., 2017; Jojic et al., 2018), it is currently unclear whether the expression of adapting genes can appear in tissues where, and/or at times when, they are not normally expressed.

In this study, we provide examples of transcriptional adaptation in Caenorhabditis elegans and show the ectopic expression of an extrachromosomal reporter in a tissue where it is not normally expressed. In addition, we analyze these transcriptional adaptation models after RNAi-mediated knockdown of different genes involved in RNA metabolism and find that the upregulation of the adapting genes requires factors involved in the maturation and transport into the nucleus of small RNAs (sRNAs).

Results

Examples of transcriptional adaptation in C. elegans

Actins are essential structural components of eukaryotic cells as they mediate a wide range of cellular processes (Pollard and Cooper, 1986). Actin genes are often present in multiple copies in higher eukaryotic genomes and hints of transcriptional adaptation modulating their expression have been reported. For example, mouse embryonic fibroblasts (MEFs) mutant for β-Actin (Actb) display increased expression of other Actins including α- and γ-Actin (ACTA and ACTG1/2) as measured by Western blots (Tondeleir et al., 2012). Similarly, Actg1 knockout, but not knockdown, in MEFs leads to an increase in Acta mRNA levels (Patrinostro et al., 2017), and zebrafish actc1b mutants exhibit mild muscle defects because of the transcriptional upregulation of actc1a (Sztal et al., 2018). Furthermore, Actg1 mutant MEFs and Actb mutant mouse embryonic stem cells (mESCs) display increased mRNA levels of Actg2 and Actg1, respectively, and this upregulation is triggered not by the loss of protein function but by mutant mRNA decay (El-Brolosy et al., 2019). Thus, we decided to investigate actin genes in C. elegans to test for transcriptional adaptation. The C. elegans genome contains five actin genes which display high similarity in their DNA and protein sequences (MacQueen et al., 2005). We started by analyzing several mutant alleles for act-1, act-2, act-3 and act-5, and determined mutant transcript levels. We found significantly reduced act-5 mRNA levels in act-5(dt2019) mutants compared to wild type (Figure 1A), likely caused by nonsense-mediated decay (NMD) due to a premature termination codon (ptc) in the first exon (Figure 1A, Figure 1—figure supplement 1). Most mutant alleles of act-5 cause severe phenotypes including lethality (Estes et al., 2011; MacQueen et al., 2005), sterility (Cui et al., 2004), and paralysis (Etheridge et al., 2015). However, the act-5(dt2019) allele, hereafter referred to as act-5(ptc), does not exhibit any obvious phenotype (MacQueen et al., 2005), an observation we confirmed. We analyzed the mRNA levels of all actin genes in act-5(ptc) mutants (Figure 1A, Figure 1—figure supplement 2), and observed the upregulation of act-3 mRNA and pre-mRNA (Figure 1B, Figure 1—figure supplement 3), consistent with a transcriptional adaptation response.

Figure 1.

mRNA levels of act-5 and act-3 in WT and mutant alleles.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in WT and act-5(ptc), act-5(Δ1), and act-5(Δ2) mutants. act-3 mRNA levels are upregulated when act-5 mutant mRNA levels are reduced (i.e., only in the act-5(ptc) allele). WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 1—source data 1. Two-tailed Student’s t-test was used to calculate P values.

(A) Schematic of all known act-5 isoforms and alleles used in this study. Black boxes indicate the deletion and nonsense alleles used in this study; black arrows point to the position of the qPCR primers. The act-5(ok1397) isoform was only identified in the act-5(Δ2) allele. (B) Partial sequence of act-5 (WT, ptc and Δ1 alleles). The PTC in the act-5(ptc) allele is 264 nucleotides from the next exon-intron junction and 888 nucleotides from the stop codon. Red indicates the mutation; PTC is underlined; ‘nnn’ and ‘---’ indicate deleted nucleotides in the act-5(Δ1) allele. The dt2017 deletion leads to a PTC which is located in the last exon (153 bases from the stop codon) and is thus predicted not to trigger NMD (Kashima et al., 2010; Lindeboom et al., 2016). act-5(dt2019) = act-5(ptc); act-5(dt2017) = act-5(Δ1); act-5(ok1397) = act-5(Δ2).

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 1—source data 2. Two-tailed Student’s t-test was used to calculate P values.

(A) qPCR analysis of act-3 pre-mRNA levels in WT and act-5(ptc) mutants. (B) Average dCt values from qPCR analysis of act-3 pre-mRNA levels in WT and act-5(ptc) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.. Two-tailed Student’s t-test was used to calculate P values.

Figure 1—figure supplement 1.

Organization of act-5 locus.

(A) Schematic of all known act-5 isoforms and alleles used in this study. Black boxes indicate the deletion and nonsense alleles used in this study; black arrows point to the position of the qPCR primers. The act-5(ok1397) isoform was only identified in the act-5(Δ2) allele. (B) Partial sequence of act-5 (WT, ptc and Δ1 alleles). The PTC in the act-5(ptc) allele is 264 nucleotides from the next exon-intron junction and 888 nucleotides from the stop codon. Red indicates the mutation; PTC is underlined; ‘nnn’ and ‘---’ indicate deleted nucleotides in the act-5(Δ1) allele. The dt2017 deletion leads to a PTC which is located in the last exon (153 bases from the stop codon) and is thus predicted not to trigger NMD (Kashima et al., 2010; Lindeboom et al., 2016). act-5(dt2019) = act-5(ptc); act-5(dt2017) = act-5(Δ1); act-5(ok1397) = act-5(Δ2).

Figure 1—figure supplement 2.

mRNA levels of act-1 (A), act-2 (B) and act-4 (C) in WT and act-5(ptc) mutants.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 1—source data 2. Two-tailed Student’s t-test was used to calculate P values.

Figure 1—figure supplement 3.

Pre-mRNA levels of act-3 in WT and act-5(ptc) mutants.

(A) qPCR analysis of act-3 pre-mRNA levels in WT and act-5(ptc) mutants. (B) Average dCt values from qPCR analysis of act-3 pre-mRNA levels in WT and act-5(ptc) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.. Two-tailed Student’s t-test was used to calculate P values.

We also examined the act-5(dt2017) partial deletion allele, hereafter referred to as act-5(Δ1), (Figure 1—figure supplement 1) and found no significant change in act-5 mRNA levels in homozygous mutants compared to wild type (Figure 1A). Notably, act-3 mRNA levels in act-5(Δ1) mutants were not changed compared to wild type (Figure 1B). To further test whether act-3 upregulation in act-5 mutants represents a model of transcriptional adaptation, we analyzed another act-5 deletion allele (ok1397) (Estes et al., 2011), hereafter referred to as act-5(Δ2). The ok1397 deletion removes part of the promoter region and the first two exons (Figure 1—figure supplement 1). We examined this allele for the presence of any transcripts and identified a new isoform which is present in mutants but not in wild type (see Materials and methods) and consists of only 3’ sequence (Figure 1—figure supplement 1). As with the act-5(Δ1) deletion allele, we found no changes in act-5 or act-3 mRNA levels in act-5(Δ2) mutants compared to wild type (Figure 1B).

In multicellular organisms, paralogous genes are often expressed in distinct spatiotemporal patterns, an indication of subfunctionalization (Guschanski et al., 2017). For example, in C. elegans, act-3 is expressed in the pharynx (Hunt-Newbury et al., 2007) while act-5 is expressed in intestinal cells (MacQueen et al., 2005). The models of transcriptional adaptation suggest a cell-autonomous mechanism, that is the mutant mRNA can cell-autonomously trigger transcriptional adaptation. In order to test this hypothesis, we generated transcriptional reporter constructs with the act-3 or act-5 promoter region driving the expression of a red florescent protein gene (Merzlyak et al., 2007). As expected, we observed expression of the extrachromosomal act-5p::rfp transgene in the intestine in wild-type animals (Figure 2A) as well as in act-5(ptc) mutants (Figure 2—figure supplement 1). Likewise, expression of the extrachromosomal act-3p::rfp transgene was only observed in the pharynx in wild-type animals (Figure 2B), consistent with previous studies (Hunt-Newbury et al., 2007). However, extrachromosomal act-3p::rfp expression was also observed in the intestine in act-5(ptc) mutants (Figure 2C), consistent with transcriptional adaptation. In summary, we saw upregulation of expression from a synthetic and extrachromosomal act-3 promoter in tissues where act-5 is expressed, supporting the model that the mutant mRNA cell-autonomously triggers transcriptional adaptation.

Figure 2.

Extrachromosomal reporter expression in WT and mutant alleles.

(A) act-5p::rfp extrachromosomal reporter expression was observed in the intestine in 153 of 300 WT animals. (B) act-3p::rfp extrachromosomal reporter expression was observed in the pharynx in 182 of 400 WT animals. (C) act-3p::rfp extrachromosomal reporter expression was observed in the pharynx and intestine in 138 of 320 act-5(ptc) mutants.

act-5p::rfp extrachromosomal reporter expression was observed in the pharynx and intestine in 148 of 300 act-5(ptc) mutants.

Figure 2—figure supplement 1.

act-5p::rfp extrachromosomal reporter expression.

act-5p::rfp extrachromosomal reporter expression was observed in the pharynx and intestine in 148 of 300 act-5(ptc) mutants.

To identify an additional example of transcriptional adaptation in C. elegans, we turned to the titin gene family (Figure 3—source data 1). Due to their size, titin genes are frequent targets of random mutagenesis, and several PTC alleles have been identified (Jorgensen and Mango, 2002; Lipinski et al., 2011). We focused on unc-89 which has many nonsense alleles that do not exhibit an obvious phenotype, potentially indicating functional compensation. We identified three different unc-89 alleles (gk469156, gk509355, gk506355) which exhibit lower levels of mutant mRNA compared to wild type (Figure 3A) and lack an obvious phenotype. Analyzing the mRNA levels of 10 titin related genes (him-4, ttn-1, ketn-1, sax-3, unc-22, unc-52, sax-7, rig-6, unc-40, and unc-73), we found that sax-3 was upregulated in all three unc-89(ptc) alleles (Figure 3—figure supplement 1), both at the mRNA (Figure 3B) and pre-mRNA (Figure 3—figure supplement 2) levels. To test whether this upregulation of sax-3 was due to transcriptional adaptation and not to the loss of UNC-89 function, we generated a 16 kb deletion (bns7000) in unc-89, hereafter referred to as unc-89(Δ), using CRISPR/Cas9 genome editing. This deletion removes part of the promoter region and the first several exons (Figure 3—figure supplement 1). Hence, most unc-89 isoforms are not observed in unc-89(Δ) mutants (Figure 3A). Homozygous unc-89(Δ) worms are maternal-effect sterile and exhibit growth defects, phenotypes not observed in the unc-89(ptc) alleles. In the RNA-less unc-89(Δ) allele, sax-3 was not upregulated (Figure 3B), indicating that sax-3 upregulation is not due to the loss of UNC-89 function and that the mutant mRNA needs to be present for the transcriptional adaptation response. Thus, sax-3 upregulation in the unc-89(ptc) alleles is a second example of transcriptional adaptation in C. elegans.

Figure 3.

mRNA levels of unc-89 and sax-3 in WT and mutant alleles.

qPCR analysis of unc-89 (C) and sax-3 (D) mRNA levels in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3), and unc-89(Δ) mutants. sax-3 mRNA levels in unc-89 alleles are upregulated when unc-89 mutant mRNA levels are reduced, except in the deletion allele. WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 3—source data 2. Two-tailed Student’s t-test was used to calculate P values.

‘-’ indicates the absence of the PTC from the isoform.

Schematic of a subset of the 16 known unc-89 isoforms as well as the deletion and nonsense alleles used in this study (black boxes). Black arrows point to the position of the qPCR primers. (B) Partial sequence of the longest unc-89 isoform with the single nucleotide change causing PTCs. Red indicates the mutations; PTCs are underlined. The distance from each PTC to the next exon-intron junction and to the stop codon is shown in Figure 3—source data 3. unc-89 () = unc-89(ptc1); unc-89 () = unc-89(ptc2); unc-89 () = unc-89(ptc3); unc-89(bns7000) = unc-89(Δ).

(A) qPCR analysis of sax-3 pre-mRNA levels in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3) mutants. (B) Average dCt values from qPCR analysis of sax-3 pre-mRNA levels in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.. Two-tailed Student’s t-test was used to calculate P values.

qPCR analysis of unc-89 (A) and sax-3 (B) mRNA levels in WT and act-5(ptc) mutants as well as of act-5 (C) and act-3 (D) in WT and unc-89(ptc1), unc-89(ptc2) and unc-89(ptc3) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.; Two-tailed Student’s t-test was used to calculate P values.

Figure 3—figure supplement 1.

Organization of unc-89 locus.

Schematic of a subset of the 16 known unc-89 isoforms as well as the deletion and nonsense alleles used in this study (black boxes). Black arrows point to the position of the qPCR primers. (B) Partial sequence of the longest unc-89 isoform with the single nucleotide change causing PTCs. Red indicates the mutations; PTCs are underlined. The distance from each PTC to the next exon-intron junction and to the stop codon is shown in Figure 3—source data 3. unc-89 () = unc-89(ptc1); unc-89 () = unc-89(ptc2); unc-89 () = unc-89(ptc3); unc-89(bns7000) = unc-89(Δ).

Figure 3—figure supplement 2.

Pre-mRNA levels of sax-3 in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3) mutants.

(A) qPCR analysis of sax-3 pre-mRNA levels in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3) mutants. (B) Average dCt values from qPCR analysis of sax-3 pre-mRNA levels in WT and unc-89(ptc1), unc-89(ptc2), unc-89(ptc3) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.. Two-tailed Student’s t-test was used to calculate P values.

Distance, in nucleotides, from each PTC to the next exon-intron junction and to the stop codon in each unc-89 isoform in the unc-89(ptc1), unc-89(ptc2), and unc-89(ptc3) alleles.

To test whether the observed changes in gene expression in act-5(ptc) and unc-89(ptc) mutants were specific, we measured unc-89 and sax-3 expression in act-5(ptc) mutants as well as act-5 and act-3 expression in unc-89(ptc) mutants. We observed no significant differences (Figure 3—figure supplement 3), suggesting that there is specificity to the gene expression changes.

Figure 3—figure supplement 3.

mRNA levels in WT and mutant alleles.

qPCR analysis of unc-89 (A) and sax-3 (B) mRNA levels in WT and act-5(ptc) mutants as well as of act-5 (C) and act-3 (D) in WT and unc-89(ptc1), unc-89(ptc2) and unc-89(ptc3) mutants. WT expression levels are set at 1. Data are mean ± S.E.M.; Two-tailed Student’s t-test was used to calculate P values.

Identifying additional regulators of transcriptional adaptation

The mutant mRNA has been reported to activate transcriptional adaptation in zebrafish embryos and mouse cell lines (El-Brolosy et al., 2019; Ma et al., 2019). In order to identify additional factors involved in transcriptional adaptation, we performed a candidate RNA interference (RNAi) screen, focusing on genes involved in RNA metabolism (Figure 4—source data 1). We knocked down genes involved in mRNA processing including splicing and nonsense-mediated decay, as well as other genes involved in small RNA synthesis and maturation. We measured the mRNA levels of the mutant and adapting genes in order to position RNAi candidates upstream or downstream of mRNA decay. If the gene targeted by RNAi is required for mutant mRNA decay, we expect to see the mRNA levels of the mutant and adapting genes to be similar to wild-type levels. If the gene targeted by RNAi is involved in transcriptional adaptation downstream of mutant mRNA decay, we expect to see the levels of mutant mRNA remaining lower than in wild type, but the levels of the adapting gene’s mRNA to be similar to wild-type levels. Finally, if the gene targeted by RNAi is not involved in transcriptional adaptation, we expect to see the levels of mutant mRNA remain lower than in wild type and the expression levels of the adapting gene to remain higher than in wild type. For example, when we knocked down drsh-1, a gene involved in miRNA biogenesis (Denli et al., 2004), we saw no significant changes in the mRNA levels of the mutant or adapting genes compared to control (Figure 4—figure supplement 1A,B,C,D), suggesting that drsh-1 is not involved in regulating transcriptional adaptation.

Figure 4—figure supplement 1.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in WT and act-5(ptc) mutants as well as of unc-89 (C) and sax-3 (D) mRNA levels in WT and unc-89(ptc) mutants upon drsh-1 RNAi-mediated knockdown by two independent clones.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

Transcriptional adaptation requires the activity of decay factors (El-Brolosy et al., 2019; Ma et al., 2019), and UPF1, SMG6, and XRN1 were reported to be differentially required in various zebrafish embryo and mouse cell line models of transcriptional adaptation (El-Brolosy et al., 2019). In order to test whether NMD factors are involved in regulating transcriptional adaptation in C. elegans, we knocked down several NMD genes including smg-2 (the C. elegans orthologue of Upf1), smg-4 (Upf3) and smg-6 (Smg6). Knockdown of smg-2 and smg-4 blocked the transcriptional adaptation response in all three unc-89(ptc) alleles but not in the act-5(ptc) allele (Figure 4). Conversely, knockdown of smg-6 blocked the transcriptional adaptation response in the act-5(ptc) allele but not in the three unc-89(ptc) alleles (Figure 4). A differential requirement for Upf1 and Smg6 between gene models was also observed in mouse cells (El-Brolosy et al., 2019).

Figure 4.

Factors regulating transcriptional adaptation identified in RNAi-mediated knockdown screen.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

As RNAi efficiency can vary in different tissues (Ratliff et al., 2006; Zhuang and Hunter, 2011), we generated double mutant strains with smg-2, smg-4, or smg-6 mutant alleles and the act-5(ptc) and unc-89(ptc) alleles to exclude the possibility of tissue-specific knockdown. Analysis of the double mutant strains confirmed the observations made in the RNAi experiments (Figure 5). For example, we found that the levels of act-5 mRNA were lower in smg-2; act-5(ptc) and smg-4; act-5(ptc) double mutants than in smg-2 and smg-4 single mutants, and that the levels of the adapting gene’s mRNA were higher (Figure 5—source data 1), further indicating that smg-2 and smg-4 are not required for transcriptional adaptation in the act-5 model. However, in smg-4; unc-89(ptc) animals, the mRNA levels of the mutant (unc-89) and adapting (sax-3) genes were similar to those in smg-4 single mutants. Furthermore, these animals exhibited a mild uncoordinated phenotype and grew slowly, suggesting a lack of functional compensation. These data further indicate that smg-4 is required for transcriptional adaptation in the unc-89 model. We could not obtain smg-6; act-5(ptc) viable mutants due to severe larval lethality, possibly as a consequence of blocking the transcriptional adaptation response, that is act-3 upregulation. smg-6; unc-89(ptc) mutants exhibited lower levels of unc-89 mRNA and higher levels of adapting gene mRNA in comparison to single smg-6 mutants (Figure 5—source data 1) similar to the observations in the RNAi experiments. Thus, there are differential requirements for decay factors in different models of transcriptional adaptation.

Figure 5.

Factors regulating transcriptional adaptation analyzed in double mutants.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in nrde-3(gg66) mutants and act-5(ptc); nrde-3(gg66) double mutants as well as of unc-89 (C) and sax-3 (D) mRNA levels in nrde-3(gg66) and unc-89(ptc); nrde-3(gg66) double mutants. Single mutant nrde-3(gg66) expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 5—source data 1. Two-tailed Student’s t-test was used to calculate P values.

Previous data indicate that the exonuclease Xrn1 is involved in regulating the transcriptional adaptation response in mouse cells (El-Brolosy et al., 2019). Therefore, we tested the role of exonucleases in transcriptional adaptation in C. elegans, specifically the exonuclease gene xrn-1 (Jones et al., 2012) and the XRN-2 partner gene paxt-1 (Miki et al., 2014) (Figure 4). We found that knocking down xrn-1 or paxt-1 led to mutant (act-5 and unc-89) mRNA levels similar to wild-type levels. Furthermore, the transcriptional adaptation response was blocked, suggesting that the degradation and processing of mutant transcripts is important to trigger transcriptional adaptation.

We next looked for additional factors required for transcriptional adaptation. Pre-mRNA splicing and NMD are closely related processes via the positioning and use of the exon junction complex (EJC) (Lejeune and Maquat, 2005; Kashima et al., 2010; Fukumura et al., 2016). SR-protein kinases (SRPK) and their substrates, serine/arginine-rich (SR) splicing factors, are key components of the splicing machinery and are well conserved across phyla (Kuroyanagi et al., 2000; Black, 2003; Galvin et al., 2011). Multiple SR proteins are components of the EJC (Singh et al., 2012), consistent with a previously suggested role of SR proteins in mRNA surveillance (Zhang and Krainer, 2004). Knocking down the SRPK gene spk-1 resulted in mutant mRNA levels similar to wild-type levels, and blocked transcriptional adaptation in all act-5 and unc-89 ptc alleles (Figure 4—figure supplement 2A,B,C,D). We also identified the SR family gene rsp-6 as a regulator of transcriptional adaptation in the act-5 model, but were unable to identify a single SR protein whose knockdown influenced the transcriptional adaptation response in the unc-89 model (Figure 4—source data 1), possibly due in part to the complexity of the unc-89 gene structure including the large number of isoforms (Tourasse et al., 2017).

Figure 4—figure supplement 2.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in WT and act-5(ptc) mutants as well as of unc-89 (C) and sax-3 (D) mRNA levels in WT and unc-89(ptc) mutants upon spk-1 RNAi-mediated knockdown by two independent clones.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

The next group of genes we targeted encode factors involved in small RNA (sRNA) biogenesis, maturation and transport into the nucleus. We tested several pathways (Figure 4—source data 1) and observed that the argonaute proteins ERGO-1 and NRDE-3, the RNA-dependent RNA polymerase RRF-3, as well as the ribonuclease DCR-1 regulate the transcriptional adaptation response downstream of mRNA decay (i.e., the mutant mRNA was still degraded but the adapting gene was not upregulated) (Figure 4, Figure 4—figure supplement 3A,B,C,D, Figure 4—source data 1). These RNAi data were confirmed by analyzing double mutants of the act-5(ptc) or unc-89(ptc) alleles with ergo-1, nrde-3, and rrf-3 (Figure 5), and all these animals exhibited phenotypes comparable to the act-5 or unc-89 deletion alleles analyzed in this study including larval lethality, slow growth and uncoordinated movements, indicating lack of functional compensation. Notably, ERGO-1, NRDE-3, RRF-3, and DCR-1 are involved in 26G RNA biogenesis (Pavelec et al., 2009; Vasale et al., 2010; Fischer, 2010; Grishok, 2013; Yvert, 2014), suggesting that 26G RNAs could play a role in transcriptional adaptation.

Figure 4—figure supplement 3.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in WT and act-5(ptc) mutants as well as of unc-89 (C) and sax-3 (D) mRNA levels in WT and unc-89(ptc) mutants upon nrde-3 RNAi-mediated knockdown by two independent clones.

WT expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 4—source data 1. Two-tailed Student’s t-test was used to calculate P values.

Together, these results indicate that mRNA decay as well as small RNA biogenesis and transport are critical in triggering transcriptional adaptation.

Discussion

Recent advances in reverse genetic tools have significantly expanded our ability to generate genetic modifications in a wide range of organisms (Housden et al., 2017). However, some engineered mutants exhibit no apparent phenotype, renewing interest in the concept of genetic robustness. Genetic compensation, and in particular transcriptional adaptation, have been proposed as a means to achieve genetic robustness upstream of protein feedback loops. Despite the potential importance of transcriptional adaptation, its underlying molecular mechanisms remain relatively unexplored. Here, we report two cases of transcriptional adaptation in C. elegans. By carrying out a small RNAi screen and a follow up analysis using double mutants, we identified several new factors that regulate transcriptional adaptation and further validated previously identified ones.

In the C. elegans act-5 model, the mutant gene and related adapting gene (act-5 and act-3, respectively) are primarily expressed in distinct tissues. However, using an extrachromosomal transcriptional reporter, we observed that in act-5(ptc) mutants, the act-3 promoter adapts to drive transcription in the primary site of act-5 expression, the intestine. As act-5(ptc) mutants do not exhibit any obvious phenotype when the transcriptional adaptation response is intact, we predict that ACT-3 and/or other proteins are able to compensate for the loss of ACT-5. Indeed, when we disrupted transcriptional adaptation, act-5(ptc) mutants did not survive. Thus, transcriptional adaptation can in some cases entail the change in the pattern of expression of related gene(s) and suppress phenotypes that would alter the animal’s fitness.

Based on the factors identified in this study, we hypothesize that the transcriptional adaptation response consists of at least three critical processes: mutant mRNA decay, sRNA maturation and sRNA transport. In terms of mutant mRNA decay, we found that the machinery can be gene-specific. In our experiments, SMG-6 is involved in act-5(ptc) mRNA decay, while SMG-2 (UPF1) and SMG-4 (UPF3) impact unc-89(ptc) mRNA decay. Similar observations were made in mouse Actb and Rela mutant cells in which siRNA-mediated knock down of SMG6 blocked the transcriptional adaptation response in Actb mutant mESCs but had little influence on Rela mutant MEFs. Conversely, siRNA-mediated knockdown of UPF1 blocked the transcriptional adaptation response in Rela mutant MEFs but not in Actb mutant mESCs (El-Brolosy et al., 2019). Consistently, mutant mRNA decay can involve different factors in the same organism (Nickless et al., 2017), possibly due to differential expression of the decay factors. However, we cannot exclude the possibility that SMG6 could function as a decay factor independent of NMD, especially since it has been reported to have NMD-independent cleavage activity (Gehring et al., 2005; Glavan et al., 2006; Huntzinger et al., 2008; Chakrabarti et al., 2014).

Transcriptional adaptation can be triggered by the degradation products of the mutant mRNA (El-Brolosy et al., 2019), which could seed the generation of sRNAs (Mattick and Makunin, 2005; Boivin et al., 2018). We found that factors involved in sRNA maturation and transport, including RRF-3, DCR-1, ERGO-1 and NRDE-3, also regulate transcriptional adaptation. Transcriptional modulation of genes by sRNAs of approximately 20–30 nucleotides in length is a widespread and diverse feature of prokaryotes (Melamed et al., 2019) and eukaryotes (Ambros et al., 2003; Yigit et al., 2006; Hutvagner and Simard, 2008; Portnoy et al., 2011; Castel and Martienssen, 2013; Rechavi and Lev, 2017; Billmyre et al., 2019). Notably, the factors we identified are known to be involved in somatic gene regulation by sRNAs, described as the RRF-3 pathway (Gent et al., 2010). RRF-3 is an RNA-dependent RNA polymerase involved along with the DICER complex in the biogenesis of 26-nucleotide RNAs with 5' bias for guanosine monophosphate (26G-RNAs) (Han et al., 2009; Gent et al., 2010; Vasale et al., 2010). 26G sRNAs associate with the Argonaute protein ERGO-1, which is involved in the further maturation of sRNAs and is required to separate the sRNA duplex (Han et al., 2009; Gent et al., 2010; Fischer et al., 2011). Mature sRNAs interacting with Argonaute proteins can direct post-transcriptional gene silencing (Vasale et al., 2010; Phillips et al., 2012), or be transported into the nucleus (Guang et al., 2008; Buckley et al., 2012; Shirayama et al., 2012). NRDE-3 is another Argonaute protein involved in transporting sRNAs into the nucleus (Guang et al., 2008), and we found that knocking it down, and knocking it out, blocked transcriptional adaptation while not affecting mutant mRNA levels. While sRNAs are best known as repressors of gene expression, they can also function as activators (Li et al., 2006; Janowski et al., 2006; Turunen et al., 2009; Portnoy et al., 2011; Wedeles et al., 2013; Li, 2017), although the underlying mechanisms remain poorly understood (Portnoy et al., 2011). Some of these activating sRNAs interact with Argonaute proteins (Seth et al., 2013), and they can target gene regulatory sequences including promoters. Whether they can also interfere with antisense RNAs, which usually function to repress gene expression (Faghihi and Wahlestedt, 2009; Modarresi et al., 2012), is a hypothesis worth testing given our observations in zebrafish embryos and mouse cell lines (El-Brolosy et al., 2019) as well as the previously suggested role of Argonaute proteins in such a process (Ghanbarian et al., 2017).

The transcriptional adaptation factors identified here came from a candidate screen where we specifically targeted pathways involved in RNA metabolism. With this study, we have established C. elegans as a genetic model system to perform unbiased screens to help reveal further mechanisms of transcriptional adaptation, a newly uncovered phenomenon contributing to genetic robustness.

Materials and methods

Culture conditions and strains

All wild-type worms were the N2 reference strain. All C. elegans strains were kept on 6 cm plates with nematode growth medium agar and fed with a lawn of E. coli OP50 grown in 500 μl Luria broth, except for the RNAi mediated knockdown experiments where the worms were fed with E. coli expressing the respective double-stranded RNA. Cultures were maintained at 20°C. Also, to minimize the potential for laboratory evolution of the trait, a new culture of the strains was revived annually from frozen stocks. All plates with fungal or bacterial contamination were excluded from the experiments.

Synchronization of cultures for RNA isolation

Worms from healthy cultures were washed off of plates using M9 buffer and passed through a 41 µm filter (Millipore Cat. No SCNY00040) with vacuum; antibiotics (Ampicillin, Chloramphenicol) were added (50 µg/ml final concentration) to eliminate remaining food bacteria, and the worms were then incubated on a shaker at room temperature for 15 min. Worms were centrifuged at 3000 rpm for 5 min to pellet early larval stage animals. The buffer was aspirated and 1 ml of fresh buffer was added to resuspend the pellet. Samples were confirmed to be primarily L1 and L2 stage larvae by observing two 5 µl samples on a 6 cm nematode growth medium plate. Starving cultures or cultures that had more than one male were excluded from the experiments.

qPCR analysis

Total RNA from synchronized cultures or manually picked young adults was isolated using TRIzol (ambion by Takara). For reverse transcription (RT), Superscript III reverse transcriptase (Invitrogen, Cat. No: 18080–044) was used following manufacturer's instructions. We used 1–2 μg total RNA for each RT reaction. The qPCR experiments were performed on a CFX Connect Real-Time System (Biorad-Roche Diagnostics) as described previously (El-Brolosy et al., 2019). cdc-42 and Y45F10D.4 (iscu-1) were used as reference genes as described previously (Hoogewijs et al., 2008), and the Ct values ranged from 12.3 to 28.4 for cdc-42 and 11.8 to 26 for Y45F10D.4. The Ct values for all other genes were aimed to be below 30.

The following primers were used to amplify the cDNA of target genes: Y45F10D.4 (forward 5’- CGAGAACCCGCGAAATGTCGGA-3’ and reverse 5’- CGGTTGCCAGGGAAGATGAGGC-3’), cdc-42 (forward 5’-AGCCATTCTGGCCGCTCTCG-3’ and reverse 5’- GCAACCGCTTCTCGTTTGGC-3’), act-1 (forward 5’-ACGACGAGTCCGGCCCATCC-3’ and reverse 5’-GAAAGCTGGTGGTGACGATGGTT-3’), act-2 (forward 5’-GCGCAAGTACTCCGTCTGGATCG-3’ and reverse 5’- GGGTGTGAAAATCCGTAAGGCAGA-3’), act-3 (forward 5’-AAGCTCTTCGCCTTACCATTTTCTC-3’ and reverse 5’-ACAGAGCAAATTGTAGTGGGGTCTTC-3’), act-4 (forward 5’-AGAGGCTCTCTTCCAGCCATCCTTC-3’ and reverse 5’-TGATCTTGATCTTCATGGTGGATGG-3’), act-5 (forward 5’- AAGTGCGATGTCGACATCAGAAAG-3’ and reverse 5’- TAATCTTGATCTTCATTGTGCTTGG-3’), act-5d (forward 5’- AAGTGCGATGTCGACATCAGAAAG-3’ and reverse 5’- TAATCTTGATCTTCATTGTGCTCCGG-3’), unc-89 (forward 5’-AAGGCTGAACTTGTCATCGAAGGAG-3’ and reverse 5’-TCATCTCCACAACATTACCCTCGTG-3’), sax-3 (forward 5’-TGCCGTTTGTCCCGTAACAACTATG-3’ and reverse 5’-ATCTTCTGAAGCTGACGGGGAGAAC-3’), act-3 pre-mRNA (forward 5’-TTTTTCAGAACCATGAAGATCA-3’ and reverse 5’-GAAAATGGTAAGGCGAAGAGC-3’), sax-3 pre-mRNA (forward 5’-TGTAAACCGCACTGCACAAT-3’ and reverse 5’-TCCACCAAGAGCCTGAAAAC-3’). PCR efficiency was determined using external standards on plasmid mini-preparation of cloned PCR products. Expression levels were analyzed by basic relative quantification. qPCR data are based on three biological replicates and three technical replicates for each biological replicate.

Rapid amplification of cDNA ends (RACE)

Total RNA from manually picked young wild-type and act-5(Δ2) mutant adults was isolated using TRIzol (ambion by Takara). 5’ and 3’ RACE ready cDNA was synthesized by reverse transcription PCR using a SMARTer RACE cDNA Amplification Kit following manufacturer's protocol (Cat. N. 634860, Takara). PCRs were performed using an Advantage 2 PCR kit (Cat. N. 639207, Takara). The following gene-specific primers and nested gene-specific primers were used to amplify 3’ and 5’ cDNA ends: act5GSP2 (5’-ACCACCGGAATCGTTTTGGACACCGGAG-3’), act5NGSP2 (5’-GAAGGATATGCCCTCCCACATGCCATCC-3’), act5GSP1 (5'-AAAAATCAGCTTAGAAGCACTTTCGGTG-3’), act5NGSP1 (5’-TCGATGGGCCGGACTCGTCGTACTCCTG-3’), unc89GSP2 (5’-TTTGGTACCATTTGTATAGAGGCGAGTG-3’), unc89NGSP2 (5’-TTCTGAACTGGACAAATCTTGCTTTTCG-3), unc89N1GSP2 (5’- ACTTTCCAGTATCTCCTGGATGTTGCTTC-3’), and unc89N2GSP2 (5’- TTTGAATACTTTTTGATGAACCGTGTGC-3’). RACE experiment revealed an isoform with an alternative start which is present only in act-5(Δ2) mutants (Figure 1—figure supplement 1A). This new isoform is not affected by the large deletion, and thus the corresponding mRNA is not degraded (Figure 1A).

Plasmid construction and genetic transformation

To study the expression of act-5, we generated a reporter construct with an act-5 promoter region (2.5 kb from III:13606066 to 13608569) fused to turboRFP in a pUC19 vector. Similarly, a pUC19 vector containing turboRFP was fused with an act-3 promoter region (4.5 kb from V:11073234 to 11077791). The germ line of wild-type animals was injected with the generated plasmids (10 ng ul−1). The transgenic lines were subsequently crossed with act-5(ptc) mutants to transfer the extrachromosomal array to the mutant background.

Confocal microscopy

A Zeiss LSM 700 confocal microscope was used to image adult worms.

RNA interference mediated knockdown

RNAi was performed by feeding double-stranded RNA-expressing bacteria at 25°C from the early larval stage through adulthood (60–75 hr) as previously described (Fraser et al., 2000). For the genes whose knockdown from an early larval stage caused lethality or sterility, we started the RNAi treatment at later stages (L4, adult). Also, for some clones (mv_R05D11.6, sjj2_R05D11.6, sjj2_Y39G8C.1), we diluted the double-stranded RNA-expressing bacteria with empty vector (L4440)-containing bacteria, in order to obtain milder effects. RNAi constructs were obtained from available libraries (Source BioScience) and verified by sequencing. RNAi clones used in this study are listed in the key resources table.

CRISPR/Cas9 induced mutations

To generate the CRISPR/Cas9-induced unc-89 deletion allele (bns7000), two sgRNAs (final concentration 4 uM each) were injected with Cas9 protein (0.35 ug/ul), and a dpy-10 sgRNA (2.5 uM) was used as a co-injection marker along with a repair oligo (PSdpy-10-PS; 0.73 uM) (Dickinson and Goldstein, 2016).

sgRNA1: 5′-GGTAGTTAGCGACCCCATGAGGG-3′.

sgRNA2: 5′-ACAGACTGGTAAACAAACGAGGG-3′

The following primers were used for genotyping: dunc-89–1 forward (5’-ATACCACCACATGTCTCTTC-3’), dunc-89–2 forward (5’-GCTAAAAGTCAGAGTTCCAC-3’), dunc-89–3 reverse (5’- GGATGGGTTTACATAAAAT-3’), dunc-89–4 reverse (5’-TGAAAAAGAAACAACAAAA-3’), dunc-89–5 forward (5’-TAACAAAAAGCTCAAAATG-3’), dunc-89–6 reverse (5’-GGATAGATTTCTGTTGGAGA-3’). The external primers flank a 19612 bp region in wild types and amplify a 3601 bp fragment in bns7000 mutants. The internal primers with different combinations amplify 500-2600 bp products in wild types.

Double mutant analysis

All the double mutants exhibited gene expression levels as in the RNAi treated animals with one exception. act-5(ptc); nrde-3 double mutants exhibited act-5 mRNA levels as in the RNAi experiments but also some upregulation of the adapting gene, unlike what was observed in the RNAi experiments (Figures 4 and 5, Figure 5—figure supplement 1A and B). One possible explanation is related to an alternative start site of nrde-3 (Tourasse et al., 2017) which might be used only in some tissues and thus could lead to some protein function in the allele used in our study.

Figure 5—figure supplement 1.

Partial data from double mutant analysis.

qPCR analysis of act-5 (A) and act-3 (B) mRNA levels in nrde-3(gg66) mutants and act-5(ptc); nrde-3(gg66) double mutants as well as of unc-89 (C) and sax-3 (D) mRNA levels in nrde-3(gg66) and unc-89(ptc); nrde-3(gg66) double mutants. Single mutant nrde-3(gg66) expression levels are set at 1. Data are mean ± S.E.M.; average dCt values are shown in Figure 5—source data 1. Two-tailed Student’s t-test was used to calculate P values.

Statistical evaluation

To calculate the significance of the differences for the expression data, we performed two-tailed Student’s t-test. Mean ± SEM is indicated in graphs. All statistical analyses were implemented in the program Statistica v. 9. Graphs were generated in Prism5.

Gene structure visualization

The act-5 and unc-89 loci were visualized using the GSDS gene structure visualization tool (Guo et al., 2007).

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The earlier discovery of what the Stainier lab calls transcriptional adaptation helped to solve confusion about discrepancies between the effects of genetic germline mutations and transient knockdowns of genes in vertebrate. This study now shows that this remarkable phenomenon is not an oddity of vertebrates, but is also detectable in invertebrates (here, nematodes). In addition to demonstrating similar genetic requirements as in vertebrates, the authors report the involvement of factors important for small RNA maturation and transport into the nucleus, which provides new mechanistic insight into transcriptional adaptation.

Decision letter after peer review:

Thank you for submitting your article "Transcriptional adaptation in Caenorhabditis elegans" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Jessica Tyler as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Michael Patrascheck (Reviewer #1); Scott Kennedy (Reviewer #2).

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

In this manuscript the authors describe the discovery of transcriptional adaptation in C. elegans. They show for two independent case that mutations in mRNAs of either act-5 or unc-89 that destabilize the mRNA can lead to a compensatory increase in related genes. They then use a limited RNAi screen to identify components related to RNA metabolism to be required in different aspects of transcriptional adaptation.

The paper is generally interesting to read and certainly attractive to a wider audience. There are however some aspects – including some critical control experiments – that need to be addressed.

1) Any particular reasons why the strains used in Figure 2 were not used to validate some of the identified factors in Table 1? Especially the act-3 expression in the gut is a very clear and powerful phenotype. The normalization necessary for qPCR experiments does not adequately account for global RNA levels changes which can be reasonably expected because many of the factors investigated are related to RNA stability. An independent validation seems necessary for at least some factors. The gut-miss-expression phenotype of act-3 seems ideal to validate some of the RNAi clones that prevent transcriptional adaptation.

2) The authors use RNAi to implicate the NMD and nuclear RNAi RNA surveillance pathways in transcriptional activation in C. elegans. Publicly available mutant alleles need to be used to confirm a role for these pathways in transcriptional activation. The worry is that RNAi knockdown induces partial knockdowns, is differentially effective in different tissues, can trigger off-targets silencing, and might trigger indirect and pleiotropic defects due to titration of the RNAi machinery. For example it is noted that smg-2/upf1 knockdown didn't suppress NMD of the unc-89 ptc allele. SMG-2/UPF1 is an obligatory component of the NMD machinery and, therefore, a failure to block NMD after SMG-2/Upf1 RNAi is likely a consequence of partial kd- and not differential requirements for SMG-2/Upf1. This concern is fairly easy to remedy. The authors should ask if animals harboring mutations in NMD and nuclear RNAi factors fail to undergo transcriptional adaptation. qRT-PCR to measure RNA levels of act-3 or sax-3 +/- NMD/nuclear RNAi null mutations would be proper. Monitoring act-5::gfp and act-3::gfp +/- NMD/nuclear RNAi would also be proper. Finally, the authors state that act-5(ptc) and unc-89(ptc) animals do not show their expected loss-of-function phenotypes… presumably because of transcriptional upregulation of homologous genes. Therefore, loss of the adaptation machinery (NMD/nuclear RNAi) in act-5(ptc) and unc-89(ptc) animals would be expected to uncover the act-5(ptc) and unc-89(ptc) lof phenotypes.

3) The data shown in Figure 4 may not be sufficient to support the author's claims that small RNAs drive transcriptional adaptation in mammals. The paper does not absolutely need this data in order to be published. Hence, one obvious option is to just remove the data from the paper. Another option would be to soften conclusions drawn from the data. Should the data stay as is, the authors should clarify the model further in the text as well as provide additional experimental support for the model. Clarification; the authors claim that the various AGO/Dicer knockdowns do not affect expression of Rela, but the data say otherwise. All four kds appear to change Rela expression and 2-3 of these changes are statistically significant (Figure 4B). Experimental support; given how AGO-1/2/3 are thought to function, I don't understand why kd of each of the three AGOs should block transcriptional adaptation. Is the idea that all AGOs act sequentially in a linear pathway to direct adaptation? Because of this perplexing result, the authors would need to provide additional data supporting the idea that mammalian AGOs drive adaptation. The authors could show that AGO depletion using an independent approach (maybe auxin/degron or Ago1/3 knockout MEFs) also blocks transcriptional adaptation. Probably easier to just edit text or remove the data.

There are three additional items that it would be nice for the authors to address, but which are not absolutely necessary for publication. These experiments are mentioned because they are all fairly straightforward and their inclusion would make the paper much stronger.

1) It would be fairly simple to provide data suggesting that up-regulation of homologous genes during "transcriptional activation" is actually due to changes in transcription by using the RNA samples (already on hand) to conduct qRT-PCR looking at act-3/unc-89 pre-mRNA levels. If the effect is transcriptional, intron containing nascent RNAs should change predictably.

2) If the model is correct, then changes in gene expression seen in animals harboring act-5/unc-89 ptc alleles should be extremely specific; i.e. limited to just homologs (because up-regulation of homologs rescues mutant phenotypes- so pleiotropic changes in gene expression should be quite minimal). An RNA-seq experiment would be an elegant way to show that this is indeed the case.

3) CRISPR/Cas9-based gfp tagging of genes is a better and more accurate way to readout gene expression than repetitive arrays integrated into random sites in the genome. Figure 2 would be more convincing if an act-3::gfp CRISPR line were used in lieu of a transgene. Or, RNA FISH showing upregulation of the act-3 mRNA in the intestine in act-5 ptc animals would also be more convincing than the current data.

The paper is generally interesting to read and certainly attractive to a wider audience. There are however some aspects – including some critical control experiments – that need to be addressed.

1) Any particular reasons why the strains used in Figure 2 were not used to validate some of the identified factors in Table 1? Especially the act-3 expression in the gut is a very clear and powerful phenotype. The normalization necessary for qPCR experiments does not adequately account for global RNA levels changes which can be reasonably expected because many of the factors investigated are related to RNA stability. An independent validation seems necessary for at least some factors. The gut-miss-expression phenotype of act-3 seems ideal to validate some of the RNAi clones that prevent transcriptional adaptation.

Indeed, the expression of the act-3p::rfp extrachromosomal reporter in the gut of act-5(ptc) mutants is a strong phenotype and appears to be an attractive tool that could be used to confirm the factors identified in the RNAi screen. However, we observed transgenerational expression of this extrachromosomal reporter. When we outcrossed act-5(ptc); Ex[act-3p::rfp] animals to transfer the extrachromosomal reporter to a wild-type background, we observed that the WT; Ex[act-3p::rfp] animals still expressed RFP in their gut for several generations. Similarly, when we tested some of the RNAi clones with act-5(ptc); Ex[act-3p::rfp] animals, most worms still exhibited RFP expression in their gut, although some at a clearly reduced level. Thus, the results were not conclusive. In summary, although the act-5(ptc); Ex[act-3p::rfp] animals are useful to illustrate transcriptional adaptation, they cannot be used in a typical screen for genes regulating this process (RNAi or genetic).

The transgenerational inheritance phenomenon observed here, which has been previously described in C. elegans, requires further investigation in the context of transcriptional adaptation.

2) The authors use RNAi to implicate the NMD and nuclear RNAi RNA surveillance pathways in transcriptional activation in C. elegans. Publicly available mutant alleles need to be used to confirm a role for these pathways in transcriptional activation. The worry is that RNAi knockdown induces partial knockdowns, is differentially effective in different tissues, can trigger off-targets silencing, and might trigger indirect and pleiotropic defects due to titration of the RNAi machinery. For example it is noted that smg-2/upf1 knockdown didn't suppress NMD of the unc-89 ptc allele. SMG-2/UPF1 is an obligatory component of the NMD machinery and, therefore, a failure to block NMD after SMG-2/Upf1 RNAi is likely a consequence of partial kd- and not differential requirements for SMG-2/Upf1. This concern is fairly easy to remedy. The authors should ask if animals harboring mutations in NMD and nuclear RNAi factors fail to undergo transcriptional adaptation. qRT-PCR to measure RNA levels of act-3 or sax-3 +/- NMD/nuclear RNAi null mutations would be proper. Monitoring act-5::gfp and act-3::gfp +/- NMD/nuclear RNAi would also be proper. Finally, the authors state that act-5(ptc) and unc-89(ptc) animals do not show their expected loss-of-function phenotypes… presumably because of transcriptional upregulation of homologous genes. Therefore, loss of the adaptation machinery (NMD/nuclear RNAi) in act-5(ptc) and unc-89(ptc) animals would be expected to uncover the act-5(ptc) and unc-89(ptc) lof phenotypes.

We fully agree with the reviewers on these points and had been generating double mutant data to incorporate into the paper. We focused on mRNA decay and sRNA biogenesis genes and the data are shown in Table 2 and Table 2—source data 1. In general, we observed effects similar to those using RNAi. In addition, we observed that double mutants that failed to undergo transcriptional adaptation phenocopied the act-5 and unc-89 deletion alleles (i.e., those that fail to upregulate the adapting gene), further illustrating the physiological relevance of transcriptional adaptation.

3) The data shown in Figure 4 may not be sufficient to support the author's claims that small RNAs drive transcriptional adaptation in mammals. The paper does not absolutely need this data in order to be published. Hence, one obvious option is to just remove the data from the paper. Another option would be to soften conclusions drawn from the data. Should the data stay as is, the authors should clarify the model further in the text as well as provide additional experimental support for the model. Clarification; the authors claim that the various AGO/Dicer knockdowns do not affect expression of Rela, but the data say otherwise. All four kds appear to change Rela expression and 2-3 of these changes are statistically significant (Figure 4B). Experimental support; given how AGO-1/2/3 are thought to function, I don't understand why kd of each of the three AGOs should block transcriptional adaptation. Is the idea that all AGOs act sequentially in a linear pathway to direct adaptation? Because of this perplexing result, the authors would need to provide additional data supporting the idea that mammalian AGOs drive adaptation. The authors could show that AGO depletion using an independent approach (maybe auxin/degron or Ago1/3 knockout MEFs) also blocks transcriptional adaptation. Probably easier to just edit text or remove the data.

We agree with the reviewers that more work needs to be done to confirm the data shown in the original Figure 4; for this reason, we have now removed these data from the manuscript.

There are three additional items that it would be nice for the authors to address, but which are not absolutely necessary for publication. These experiments are mentioned because they are all fairly straightforward and their inclusion would make the paper much stronger.

1) It would be fairly simple to provide data suggesting that up-regulation of homologous genes during "transcriptional activation" is actually due to changes in transcription by using the RNA samples (already on hand) to conduct qRT-PCR looking at act-3/unc-89 pre-mRNA levels. If the effect is transcriptional, intron containing nascent RNAs should change predictably.

We have now performed qPCR experiments and found higher pre-mRNA levels of the adapting genes in act-5(ptc) and unc-89(ptc) mutants compared to WT, suggesting that increased transcription is causing the increase in mRNA levels.

2) If the model is correct, then changes in gene expression seen in animals harboring act-5/unc-89 ptc alleles should be extremely specific; i.e. limited to just homologs (because up-regulation of homologs rescues mutant phenotypes- so pleiotropic changes in gene expression should be quite minimal). An RNA-seq experiment would be an elegant way to show that this is indeed the case.

In our published transcriptome analyses of wild type versus mutant mouse cell lines (El-Brolosy et al., 2019), we observed the differential expression of hundreds of genes. As discussed in the paper, these gene expression changes can be grouped into at least three categories: (1) those that are a direct or indirect result of protein function loss, or forms of compensation other than transcriptional adaptation, (2) those that are a result of transcriptional adaptation, and (3) those that are an indirect result of transcriptional adaptation. Thus, it is unlikely that the adapting genes we analyzed (act-3 and sax-3) are the only genes whose expression is changed in act-5(ptc) and unc-89(ptc) mutants, respectively. To test specificity for this study, we measured unc-89 and sax-3 mRNA levels in act-5(ptc) mutants, as well as act-5 and act-3 expression levels in unc-89(ptc) mutants. We did not observe any significant difference (Figure 3—figure supplement 3), indicating that the expression changes are not genome wide.

3) CRISPR/Cas9-based gfp tagging of genes is a better and more accurate way to readout gene expression than repetitive arrays integrated into random sites in the genome. Figure 2 would be more convincing if an act-3::gfp CRISPR line were used in lieu of a transgene. Or, RNA FISH showing upregulation of the act-3 mRNA in the intestine in act-5 ptc animals would also be more convincing than the current data.

We obtained several GFP-tagged act-3 lines from Craig Mello, but so far have not observed GFP expression in the gut when crossed into the act-5(ptc) mutant background. This discrepancy could be due to differences in expression levels of the transgene vs. the endogenous gene (whole animal qPCR data indicate that the extrachromosomal transgene is upregulated 256-fold vs. 1.5-fold for the endogenous gene in the act-5(ptc) background). Given the relatively high autofluorescence of the intestine, a relatively minor increase of the endogenous gene in the GFP-tagged act-3 lines may not be visible over background. It is also possible that the insertion of GFP in the act-3 locus interferes with the transcriptional adaptation response. Thus, further work is required to obtain the optimal reagent to pursue this question.

We agree that RNA FISH would be a better approach. However, the high level of sequence similarity of act-3 with other actin genes makes it challenging to design enough act-3-specific probes required for the experiment.

Key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Gene (Caenorhabditis elegans)	act-1		CELE_T04C12.6	WormBase ID: WBGene00000063
Gene (Caenorhabditis elegans)	act-2		CELE_T04C12.5	WBGene00000064
Gene (Caenorhabditis elegans)	act-3		CELE_T04C12.4	WBGene00000065
Gene (Caenorhabditis elegans)	act-4		CELE_M03F4.2	WBGene00000066
Gene (Caenorhabditis elegans)	act-5		CELE_T25C8.2	WBGene00000067
Gene (Caenorhabditis elegans)	unc-89		CELE_C09D1.1	WBGene00006820
Gene (Caenorhabditis elegans)	sax-3		CELE_ZK377.2	WBGene00004729
Strain, strain background (C. elegans)	N2	CGC, Bristol strain		wild type
Strain, strain background (C. elegans)	IN2049	MacQueen et al., 2005		act-5(ptc); dtIs419[act-5+ rol-6(d)]
Strain, strain background (C. elegans)	IN2051	MacQueen et al., 2005		act-5(Δ1); dtIs419[act-5+ rol-6(d)]
Strain, strain background (C. elegans)	VC971	CGC, Estes et al., 2011		+/mT1; act-5(Δ2)/mT1 [dpy-10(e128)].
Strain, strain background (C. elegans)	CB4043	CGC, Hodgkin et al., 1989		smg-2(e2008);him-5(e1490)
Strain, strain background (C. elegans)	CB4355	CGC, Pulak and Anderson, 1993		smg-4(ma116);him-8(e1490)
Strain, strain background (C. elegans)	TR1396	CGC, Pulak and Anderson, 1993		smg-6(r896)
Strain, strain background (C. elegans)	YY168	CGC, Pavelec et al., 2009		ergo-1(gg100)
Strain, strain background (C. elegans)	YY158	CGC, Guang et al., 2008		nrde-3(gg66)
Strain, strain background (C. elegans)	YY13	CGC, Pavelec et al., 2009		rrf-3(mg373)
Strain, strain background (C. elegans)	DYS0005	This study, crossed IN2049 to N2		act-5(ptc)
Strain, strain background (C. elegans)	DYS0004	This study, crossed IN2049 to N2		+/act-5(Δ1)
Strain, strain background (C. elegans)	DYS0012	This study, injected in N2		Ex[act-5p::RFP]
Strain, strain background (C. elegans)	DYS0014	This study, injected in N2		Ex[act-3p::RFP]
Strain, strain background (C. elegans)	DYS0015	This study, crossed DYS0014 to DYS0004		act-5(ptc);Ex[act-3p::RFP]
Strain, strain background (C. elegans)	DYS0042	This study, crossed DYS0012 to DYS0005		act-5(ptc);Ex[act-5p::RFP]
Strain, strain background (C. elegans)	VC40114	CGC, Million Mutation Project		unc-89(ptc1)
Strain, strain background (C. elegans)	VC40193	CGC, Million Mutation Project		unc-89(ptc2)
Strain, strain background (C. elegans)	VC40199	CGC, Million Mutation Project		unc-89(ptc3)
Strain, strain background (C. elegans)	DYS0028	This study, crossed VC40114 to N2		unc-89(ptc1)
Strain, strain background (C. elegans)	DYS0030	This study, crossed VC40193 to N2		unc-89(ptc2)
Strain, strain background (C. elegans)	DYS0031	This study, crossed VC40199 to N2		unc-89(ptc3)
Strain, strain background (C. elegans)	DYS0037	This study, induced by CRISPR/Cas9		unc-89(Δ)
Strain, strain background (C. elegans)	DYS0008	This study, crossed DYS0005 to CB4043		smg-2(e2008); act-5(ptc)
Strain, strain background (C. elegans)	DYS0057	This study, crossed DYS0005 to CB4355		act-5(ptc); smg-4(ma116)
Strain, strain background (C. elegans)	DYS0047	This study, crossed DYS0028 to CB4355		unc-89(ptc1); smg-4(ma116)
Strain, strain background (C. elegans)	DYS0048	This study, crossed DYS0030 to CB4355		unc-89(ptc2); smg-4(ma116)
Strain, strain background (C. elegans)	DYS0050	This study, crossed DYS0031 to CB4355		unc-89(ptc3); smg-4(ma116)
Strain, strain background (C. elegans)	DYS0053	This study, crossed DYS0028 to TR1396		unc-89(ptc1); smg-6(r896)
Strain, strain background (C. elegans)	DYS0055	This study, crossed DYS0030 to TR1396		unc-89(ptc2); smg-6(r896)
Strain, strain background (C. elegans)	DYS0056	This study, crossed DYS0031 to TR1396		unc-89(ptc3); smg-6(r896)
Strain, strain background (C. elegans)	DYS0010	This study, crossed DYS0005 to YY168		act-5(ptc); ergo-1(gg100)
Strain, strain background (C. elegans)	DYS0054	This study, crossed DYS0028 to YY168		unc-89(ptc1); ergo-1(gg100)
Strain, strain background (C. elegans)	DYS0051	This study, crossed DYS0030 to YY168		unc-89(ptc2); ergo-1(gg100)
Strain, strain background (C. elegans)	DYS0052	This study, crossed DYS0031 to YY168		unc-89(ptc3); ergo-1(gg100)
Strain, strain background (C. elegans)	DYS0045	This study, crossed DYS0005 to YY158		act-5(ptc); nrde-3(gg66)
Strain, strain background (C. elegans)	DYS0065	This study, crossed DYS0028 to YY158		unc-89(ptc1); nrde-3(gg66)
Strain, strain background (C. elegans)	DYS0072	This study, crossed DYS0030 to YY158		unc-89(ptc2); nrde-3(gg66)
Strain, strain background (C. elegans)	DYS0066	This study, crossed DYS0031 to YY158		unc-89(ptc3); nrde-3(gg66)
Strain, strain background (C. elegans)	DYS0046	This study, crossed DYS0005 to YY13		rrf-3(mg373); act-5(ptc)
Strain, strain background (C. elegans)	DYS0070	This study, crossed DYS0028 to YY13		unc-89(ptc1); rrf-3(mg373)
Strain, strain background (C. elegans)	DYS0062	This study, crossed DYS0030 to YY13		unc-89(ptc2); rrf-3(mg373)
Strain, strain background (C. elegans)	DYS0063	This study, crossed DYS0031 to YY13		unc-89(ptc3); rrf-3(mg373)
Commercial assay or kit	In-Fusion HD Cloning	Clontech	Clontech:639647
Commercial assay or kit	Superscript III reverse transcriptase	Takara	Cat. No: 18080–044
Commercial assay or kit	SMARTer RACE cDNA Amplification Kit	Takara	Cat. N. 634860
Commercial assay or kit	Advantage 2 PCR kit	Takara	Cat. N. 639207
RNAi construct	mv_C18D11.4	BioScience		rsp-8
RNAi construct	sjj2_C18D11.4	BioScience		rsp-8
RNAi constructs	mv_C33H5.12	BioScience		rsp-6
RNAi constructs	sjj2_C33H5.12	BioScience		rsp-6
RNAi constructs	mv_W02B12.3	BioScience		rsp-1
RNAi constructs	sjj2_W02B12.3	BioScience		rsp-1
RNAi constructs	mv_D2089.1	BioScience		rsp-7
RNAi constructs	sjj2_D2089.1	BioScience		rsp-7
RNAi constructs	mv_B0464.5	BioScience		spk-1
RNAi constructs	sjj2_B0464.5	BioScience		spk-1
RNAi constructs	mv_R05D11.6	BioScience		paxt-1
RNAi constructs	sjj2_R05D11.6	BioScience		paxt-1
RNAi constructs	mv_F43E2.8	BioScience		hsp-4
RNAi constructs	sjj2_F43E2.8	BioScience		hsp-4
RNAi constructs	sjj2_Y39G8C.1	BioScience		xrn-1
RNAi constructs	mv_Y48G8AL.6	BioScience		smg-2
RNAi constructs	sjj2_Y48G8AL.6	BioScience		smg-2
RNAi constructs	sjj2_F46B6.3	BioScience		smg-4
RNAi constructs	mv_Y54F10AL.2	BioScience		smg-6
RNAi constructs	sjj2_Y54F10AL.2	BioScience		smg-6
RNAi constructs	mv_F26B1.2	BioScience		hrpk-1
RNAi constructs	sjj2_F26B1.2	BioScience		hrpk-1
RNAi constructs	mv_F26E4.10	BioScience		drsh-1
RNAi constructs	sjj2_F26E4.10	BioScience		drsh-1
RNAi constructs	mv_T22A3.5	BioScience		pash-1
RNAi constructs	sjj2_T22A3.5	BioScience		pash-1
RNAi constructs	sjj2_F26A3.8	BioScience		rrf-1
RNAi constructs	mv_ R06C7.1	BioScience		wago-1
RNAi constructs	sjj2_ R06C7.1	BioScience		wago-1
RNAi constructs	mv_F58G1.1	BioScience		wago-4
RNAi constructs	sjj2_F58G1.1	BioScience		wago-4
RNAi constructs	sjj2_F10B5.7	BioScience		rrf-3
RNAi constructs	mv_M88.5	BioScience		zbp-1
RNAi constructs	sjj2_M88.5	BioScience		zbp-1
RNAi constructs	sjj2_K12H4.8	BioScience		dcr-1
RNAi constructs	mv_T20G5.11	BioScience		rde-4
RNAi constructs	sjj2_T20G5.11	BioScience		rde-4
RNAi constructs	mv_F36H1.2	BioScience		kdin-1
RNAi constructs	mv_K12B6.1	BioScience		sago-1
RNAi constructs	sjj2_K12B6.1	BioScience		sago-1
RNAi constructs	mv_K08H10.7	BioScience		rde-1
RNAi constructs	sjj2_K08H10.7	BioScience		rde-1
RNAi constructs	sjj2_R09A1.1	BioScience		ergo-1
RNAi constructs	mv_R04A9.2	BioScience		nrde-3
RNAi constructs	sjj2_R04A9.2	BioScience		nrde-3