Genome engineering to create targeted gene deletions, mutations, reporter constructs, and epitope-tagged proteins is a key strategy for the mechanistic dissection of almost any biological process. Budding yeast was the first eukaryotic organism in which this became highly facile, thanks to the development of a one-step PCR-based editing strategy, frequently used with a shared toolkit of selection cassettes (Longtine et al., 1998; Wach et al., 1997; Wach et al., 1994; Baudin et al., 1993; Lorenz et al., 1995). These tools became common, routinely used by thousands of labs as well as to enable large global endeavors, such as creation of the yeast deletion collection (Giaever et al., 2002). Because of the high fidelity, ease of use, and rapid nature of this selection-mediated strategy, it has remained commonplace despite the development of methods allowing non-selection-marked mutations that include CRISPR/Cas9-based editing (Jinek et al., 2012; Güldener et al., 1996; Gray et al., 2005).

While selection-mediated editing strategies have been used to advance countless discoveries, their utility relies on the assumption that insertion of a selection cassette at one locus will not disrupt the expression of neighboring genes. However, analyses of mutant phenotypes detected in global studies of the collection of strains in which each non-essential yeast ORF is replaced with a kanMX cassette revealed effects that appeared to result from neighboring gene mis-regulation, termed ‘the neighboring gene effect’ (NGE) (Ben-Shitrit et al., 2012; Atias et al., 2016; Egorov et al., 2021). This is caused by overlap of the deleted ORF with a regulatory region for the neighboring gene in a subset of cases, as just one of several problems that resulted from the large-scale nature of the effort required to create the deletion collection (Ben-Shitrit et al., 2012; Hughes et al., 2000; Teng et al., 2013). However, most cases of the NGE remain unexplained, and this lack of mechanistic understanding makes it unclear how prevalent this effect is in traditional small-scale laboratory studies, and how to prevent it.

It is now understood that genomic loci can have complex and linked transcriptional outputs, even in yeast. For example, activity from one transcription start site (TSS) can interfere with the output of nearby TSSs in an adjacent sense or antisense configuration, and that most, if not all, promoters are bidirectionally active (Teodorovic et al., 2007; Neil et al., 2009; Preker et al., 2008; Core et al., 2008; Xu et al., 2009; Seila et al., 2008; Churchman and Weissman, 2011; Jin et al., 2017). Transcription interference is seen in diverse organisms, including yeast and human cells, and can occur between TSSs driving coding or non-coding RNAs and controlling both transcript isoform identity and transcript levels (Chia et al., 2017; Hirschman et al., 1988; Martens et al., 2004; van Werven et al., 2012; Hongay et al., 2006; Hausler and Somerville, 1979; Adhya and Gottesman, 1982; Proudfoot, 1986; Boussadia et al., 1997; Emerman and Temin, 1984; Corbin and Maniatis, 1989; Struhl, 1985). It has also been shown that more than one TSS is often present at a single locus, even in the simple budding yeast (Pelechano et al., 2013).

Transcription interference between TSSs within the same genomic locus is an important feature of a naturally occurring type of regulation dependent on long undecoded transcript isoforms (LUTIs; Chia et al., 2017; Chen et al., 2017). For genes regulated by this strategy, 5′ extended LUTIs are transcribed in place of canonical mRNAs to temporally downregulate protein synthesis (Chia et al., 2017; Chen et al., 2017; Cheng et al., 2018; Van Dalfsen et al., 2018). Here, use of an upstream alternate TSS represses transcription from the canonical TSS in cis via transcriptional interference (Chia et al., 2017). In concert, translation of the main ORF-encoded protein products from LUTIs are repressed by translation of competitive AUG-initiated upstream ORFs (uORFs) (Chen et al., 2017; Wethmar, 2014; Barbosa et al., 2013; Hinnebusch et al., 2016; Law et al., 2005; Brar et al., 2012; Cheng et al., 2018). Natural LUTI-based regulation has been shown to modulate gene expression for many genes during meiosis and the unfolded protein response in yeast (Chen et al., 2017; Cheng et al., 2018; Van Dalfsen et al., 2018). Furthermore, features of LUTI-based regulation are highly conserved across eukaryotes, and evidence suggests the presence of this regulation in more complex species, such as humans and algae (Hollerer et al., 2019; Sehgal et al., 2008; Moseley et al., 2002). Here, we show that insertion of expression cassettes commonly used edit the genomes of yeast cells, can induce the repression of neighboring genes though synthetic and constitutive LUTI-based repression.

In particular, we report that selection-cassette replacement of the ORF for DBP1 causes mis-regulation of the adjacent gene, MRP51, despite no changes to the MRP51 coding or regulatory regions, as a result of synthetic LUTI-based repression. All phenotypes we observed in cells with a cassette-mediated deletion of the DBP1 ORF were caused by mis-regulation of MRP51, rather than loss of DBP1, consistent with an NGE. We leveraged the strong off-target effects observed with cassette-mediated replacement of DBP1 to interrogate the mechanism driving this phenomenon. We find that cassette-mediated off-target effects were general to all expression cassettes that we inserted at the DBP1 locus and are driven by the bidirectional cassette promoter activity now understood to be inherent to eukaryotic promoters.

These data point to an undesirable feature of expression cassette insertion that is currently being overlooked in the design of genome-editing approaches. While we find that stable cassette-driven divergent transcripts were detected in ~30% of cassette-inserted loci, all cassette-inserted loci have detectable divergent transcripts in cells lacking nuclear exosome-mediated RNA decay. Thus, our data suggest that divergent transcription results ubiquitously from expression cassette insertion, but features—including transcription termination sequences adjacent to cassette insertion—can naturally mitigate off-target neighboring gene mis-regulation. Because it is difficult to predict from sequence information alone whether a locus will be sensitive to neighboring gene mis-regulation, we designed improved expression cassettes containing an additional terminator sequence flanking the promoter, which prevents neighboring gene disruption. Use of such cassettes should improve the specificity of engineered mutant strains for future studies. More broadly, our study uncovers a mechanism by which genome engineering can drive neighboring gene mis-expression, and that warrants consideration in the design of future studies in yeast, as well as other eukaryotes.

In this study, we provide evidence that genome-inserted expression cassettes drive bidirectional transcription and the production of divergent transcripts that can repress neighboring genes in yeast, resulting in potent off-target effects, even in carefully designed single-gene studies. Our data also provide a mechanism for the mysterious NGE, identified in analyses of yeast deletion collection studies, and clarify why some loci are susceptible to off-target effects, while some are immune (Ben-Shitrit et al., 2012; Atias et al., 2016; Egorov et al., 2021; Makeeva et al., 2019). We find that expression cassette insertion at DBP1 reduces the translation of neighboring gene MRP51 through a divergent LUTI driven by the bidirectional activity of cassette promoters (Neil et al., 2009; Xu et al., 2009; Churchman and Weissman, 2011). Production of this divergent transcript drives both transcription interference and uORF-based translation repression for MRP51 (Chia et al., 2017; Chen et al., 2017; Cheng et al., 2018). Because most, if not all, eukaryotic promoters are thought to be bidirectional, these data suggest that any inserted expression cassette containing a promoter could drive neighboring off-target effects (Teodorovic et al., 2007; Neil et al., 2009; Preker et al., 2008; Core et al., 2008; Xu et al., 2009; Seila et al., 2008).

How commonly does expression cassette insertion cause confounding mutant phenotypes? Confidence in the specificity of yeast genome-editing cassettes has motivated their widespread use, and many landmark studies defining the functions of conserved proteins have been performed using this approach. It is routine for multiple loci to be disrupted within the same strain to enable analysis of genetic interactions, and global yeast deletion, hypomorphic, and epitope tag collections were created by use of the kanMX insertion cassette (Giaever et al., 2002; Winzeler et al., 1999; Giaever and Nislow, 2014; Chu and Davis, 2008; Ghaemmaghami et al., 2003; Huh et al., 2003; Breslow et al., 2008). Despite the first reports of NGEs over 10 years ago, the original yeast cassettes remain widely used (Ben-Shitrit et al., 2012). One study that analyzed mutants generated with kanMX modules estimated that replacement at as many as 20% of loci may cause neighboring gene mis-regulation, and ~10% of deletion collection strains have been estimated to demonstrate non-specific phenotypes in genetic interaction data that could be attributed to neighboring gene mis-regulation (Ben-Shitrit et al., 2012; Atias et al., 2016; Egorov et al., 2021). A recent study focused on protein quantification of each yeast deletion collection strain found that one of the most significant factors determining whether a specific protein’s abundance would change following deletion of a gene encoding another protein was whether the genes encoding the protein pair were genomic neighbors (Messner et al., 2022). It is important to note that all previously reported examples of the NGE, including those noted here, focused only on integrations of kanMX modules (Ben-Shitrit et al., 2012; Atias et al., 2016; Egorov et al., 2021; Makeeva et al., 2019; Messner et al., 2022). Our study determined that NGEs are driven by cassettes of disparate sequence. This observation allowed us to determine that the promoter bidirectionality that is inherent to eukaryotic promoters, rather than any feature of a particular cassette sequence, drives neighboring gene disruption. Thus, any expression cassette can cause disruptive NGEs, suggesting a much larger pool of potentially affected mutants than previously considered.

In fact, we found cassette-driven divergent transcripts to be produced at 100% of expression cassette-inserted loci analyzed, but they were often short and masked by exosome-mediated RNA decay. This suggests that additional features local to cassette insertion can prevent off-target phenotypes. Given that bidirectional promoter activity originating from cassettes drives neighboring gene disruption, there are two non-mutually exclusive explanations for why some loci would produce disruptive divergent transcripts and others would not. First, the different chromatin context at inserted loci may affect the bidirectional promoter activity (Churchman and Weissman, 2011; Jin et al., 2017; Struhl, 1985; Porrua and Libri, 2015; Marquardt et al., 2014). Alternatively, bidirectional activity and divergent transcript production could occur at all loci, with off-target effects being prevented in many cases by local sequence characteristics. Our data do not exclude the possibility that chromatin context is a factor, but do suggest that the role of local sequence features in determining the span of divergent transcription is important. One such feature that may naturally prevent neighboring gene disruption at cassette-inserted loci are NNS termination signals, like those known to terminate and prevent endogenously produced divergent transcripts from interfering with neighboring genes (Steinmetz et al., 2001; Schulz et al., 2013; Porrua and Libri, 2015). Both NNS termination signals and canonical coding gene transcription termination sequences were sufficient to prevent to ‘insulate’ bidirectional promoters within selection cassettes inserted at the DBP1 locus.

How do divergent transcripts from cassette-inserted loci drive off-target phenotypes? While we focused on the integrated transcription and translation regulation driven by LUTI-based disruption of MRP51 in this study, mis-regulation of adjacent genes by cassette-driven transcription interference alone may be even more frequent. LUTI-based neighboring gene disruption requires the adjacent gene to be in a ‘head-to-head’ orientation, the presence of repressive uORFs upstream of the coding sequence, and a continuous divergent transcript containing the entire ORF of the repressed gene (Chia et al., 2017; Chen et al., 2017; Cheng et al., 2018). Cassette-driven transcription interference, however, can occur regardless of the presence of repressive uORFs upstream of the neighboring coding sequence, does not require the divergent transcript to be continuous over the neighboring ORF, and can act on genes in either orientation to the inserted cassette (Chia et al., 2017; Hirschman et al., 1988; Martens et al., 2004; van Werven et al., 2012; Hongay et al., 2006; Hausler and Somerville, 1979; Adhya and Gottesman, 1982; Proudfoot, 1986; Boussadia et al., 1997; Emerman and Temin, 1984; Corbin and Maniatis, 1989). For example, in haploid cells, induction into meiosis is blocked by the expression of two separate non-coding RNAs that repress an antisense mRNA (RME2/IME4) or a downstream sense RNA (IRT1/IME1) via transcription interference (Hongay et al., 2006; van Werven et al., 2012). Transcription interference between loci can be effective over large distances, like the ~1800 nt span over which transcription of the repressive IRT1 long non-coding RNA represses IME1 (van Werven et al., 2012).

The type of cassette-mediated off-target effect in this study is difficult to detect or predict by sequence alone. No unexpected mutations exist in these cases and even mRNA measurements of neighboring genes are not diagnostic unless all possible transcript isoforms are measured from a given locus. Additionally, our data suggest that at as many as 70% of cassette-inserted loci, divergent transcripts are produced but degraded by the nuclear exosome, meaning that even the lack of an observable stable divergent transcript is not sufficient to conclude wild-type regulation from a cassette-adjacent gene because disruption can occur regardless of the transcript’s stability. Measuring translation or protein levels for adjacent genes should be diagnostic of this undesirable side effect, but ribosome profiling is not a routine experiment that is applied to most mutants, and tagging proteins can alter function and regulation. Additionally, as many genes are expressed in a condition-specific manner, measurement of the adjacent gene’s protein levels compared to wild-type may give false confidence unless the proper conditions are chosen. Consistently, we observed condition-specific manifestation of the cassette-driven divergent transcription at the DBP1/MRP51 locus. For 3/5 additional potential NGE cases with detectable aberrant transcripts in exosome proficient cells (Figure 4), adjacent genes were lowly expressed, even in wild-type cells under the conditions surveyed. Considering the number of factors affecting whether cassette adjacent genes are mis-regulated, we argue that it is currently not possible to predict when and where cassette insertion would result in mis-regulation severe enough to mislead studies.

Despite this complexity, complementation-based rescue experiments—namely single-copy insertion of the deleted gene with its entire endogenous regulatory region at another locus—allowed us to identify the unexpected regulation in the case of DBP1 disruption, and should be more commonly used to confirm phenotypes. However, for genome-wide experiments, those using cassettes for epitope tagging, or experiments in which genetic interactions between several genes are being investigated, this may not be feasible. Therefore, we created ‘insulated’ resistance cassettes containing transcription terminator sequences flanking the promoter, which prevent off-target effects and possible misinterpretation of mutant phenotypes. Other possible alternatives to avoiding these confounding off-target effects include methods that allow mutants to be isolated without expression cassette insertion. Seamless deletions such as those created with the CRISPR/Cas9 approach avoid cassette insertions, but can still result in changes to the local environment of adjacent genes, potentially resulting in off-target effects (Jinek et al., 2012; Güldener et al., 1996; Gray et al., 2005). For the case of a markerless CRISPR/Cas9-based deletion of the DBP1 ORF, we observed translation of a previously untranslated ORF within the mutant transcript containing only the DBP1 5′ and 3′ UTR (Figure 5—figure supplement 2). One could imagine that in cases like these, translation of novel ORFs could also cause misleading neomorphic mutant phenotypes, not linked to loss of function of the deleted ORF. These data highlight the point that even well controlled genomic changes can have unintended consequences on the translation of neighboring coding sequences, and emphasize the value of careful analysis of gene expression from loci neighboring those that are manipulated.

We note that two-step strategies to isolate mutants free of selection cassettes have been described but are often used the context of allowing subsequent re-use of auxotrophic markers and cassettes (Güldener et al., 1996; Gray et al., 2005). One example of this uses the Cre/loxP system to excise resistance cassettes following mutation isolation (Güldener et al., 1996). While this strategy can prevent neighboring gene disruption by cassettes, it is slower and more labor intensive than a single step process, and thus may not be commonly utilized to its full potential. For example, in two datasets we analyzed for cassette-driven divergent transcripts in strain backgrounds deficient for exosome-mediated RNA decay, cassette integrated mutants contained flanking loxP sites that were not utilized to excise the cassette (Schmidt et al., 2012). Use of a single-step ‘insulated’ cassette for genome editing maintains the efficiency of the originally designed system, while preventing the potentially catastrophic off-target effects that disruption of neighboring gene expression can cause.

While our study was limited to yeast, all the key factors behind the cassette-induced off-target effects described here are conserved in higher eukaryotes. Because of this, we hypothesize that genome-editing-driven transcriptional interference, including LUTI-based repression, is likely to occur in other organisms. It is now understood that promoters in yeast, mammals, and other eukaryotes are inherently bidirectional, but that evolution of local sequences to can limit transcription interference of adjacent genes by transcription termination of divergent transcripts within a short distance of their initiation (Teodorovic et al., 2007; Neil et al., 2009; Seila et al., 2008; Churchman and Weissman, 2011; Jin et al., 2017; Schulz et al., 2013; Arnone, 2020; Villa et al., 2020; Ha et al., 2022; Ntini et al., 2013; Almada et al., 2013; Core et al., 2008; Xu et al., 2009). Although NNS-based termination does not prevent production of long stable divergent transcripts from bidirectional promoters in mammals, analogous evolution of early polyadenylation sites appears to have achieved the same result (Schulz et al., 2013; Ntini et al., 2013; Almada et al., 2013; Porrua and Libri, 2015). In both cases, the local context of promoters defines the degree to which they produce divergent transcripts with the ability to interfere with neighboring gene expression.

The bidirectional nature of commonly used expression cassette promoters, including pTEF in yeast, and the CMV, eEF1α, and SV40 promoters in mammalian systems, has long been known (Curtin et al., 2008; Gidoni et al., 1985). However, consequences of this for their modular use in expression constructs do not yet seem to be widely considered. Early studies that established ‘minimal’ promoters for modular use in expression cassettes explicitly served to separate the transcript-generating feature of promoter regions from their native sequence context. Because local sequence features outside of these minimal promoter regions can limit the effects of divergent transcription, and insertion of a terminator flanking inserted promoter sequences can serve a similar function, our study suggests this simple fix to prevent cassette-based off-target effects in future studies. Our study also emphasizes the need for proper controls to establish causality of observed phenotypes to the intended mutation, even when using precision genome engineering approaches in yeast and other organisms.

mRNA sequencing and ribosome profiling data are available at NCBI GEO, with accession numbers GSE207267 and GSE207189. Mass spectrometry data are available at MassIVE, with accession number MSV000089724.

1. 1. Powers E
   2. Brar G

   (2022) NCBI Gene Expression Omnibus

   ID GSE207267. Loss of DBP1 during meiosis in yeast.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE207267
2. 1. Powers E
   2. Brar G

   (2022) NCBI Gene Expression Omnibus

   ID GSE207189. Ribosome profiling and mRNA seq demonstrate that insertion of common resistance cassettes in yeast drives aberrant transcription.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE207189
3. 1. Jovanovic M
   2. Kim Kim J

   (2022) MSV000089724

   ID MassIVE. Use of a ubiquitous gene-editing tool in budding yeast causes off-target repression of neighboring gene protein synthesis.

   https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=972212bec77b48faab383a6c5c2b9310

1. 1. Brar GA
   2. Cheng Z

   (2018) NCBI Gene Expression Omnibus

   ID GSE121189. Small and large ribosomal subunit deficiencies lead to distinct gene expression signatures that reflect cellular growth rate.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE121189
2. 1. Sen ND
   2. Hinnebusch AG

   (2018) NCBI Gene Expression Omnibus

   ID GSE111255. DBP1 in budding yeast.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE111255

1. 1. Adhya S
   2. Gottesman M

   (1982) Promoter occlusion: transcription through a promoter may inhibit its activity

   Cell 29:939–944.

   https://doi.org/10.1016/0092-8674(82)90456-1

   • PubMed
   • Google Scholar
2. 1. Almada AE
   2. Wu X
   3. Kriz AJ
   4. Burge CB
   5. Sharp PA

   (2013) Promoter directionality is controlled by U1 snRNP and polyadenylation signals

   Nature 499:360–363.

   https://doi.org/10.1038/nature12349

   • PubMed
   • Google Scholar
3. 1. Arnone JT

   (2020) Genomic considerations for the modification of Saccharomyces cerevisiae for biofuel and metabolite biosynthesis

   Microorganisms 8:321.

   https://doi.org/10.3390/microorganisms8030321

   • PubMed
   • Google Scholar
4. 1. Atias N
   2. Kupiec M
   3. Sharan R

   (2016) Systematic identification and correction of annotation errors in the genetic interaction map of Saccharomyces cerevisiae

   Nucleic Acids Research 44:e50.

   https://doi.org/10.1093/nar/gkv1284

   • PubMed
   • Google Scholar
5. 1. Barbosa C
   2. Peixeiro I
   3. Romão L

   (2013) Gene expression regulation by upstream open reading frames and human disease

   PLOS Genetics 9:e1003529.

   https://doi.org/10.1371/journal.pgen.1003529

   • PubMed
   • Google Scholar
6. 1. Baudin A
   2. Ozier-Kalogeropoulos O
   3. Denouel A
   4. Lacroute F
   5. Cullin C

   (1993) A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae

   Nucleic Acids Research 21:3329–3330.

   https://doi.org/10.1093/nar/21.14.3329

   • PubMed
   • Google Scholar
7. 1. Ben-Shitrit T
   2. Yosef N
   3. Shemesh K
   4. Sharan R
   5. Ruppin E
   6. Kupiec M

   (2012) Systematic identification of gene annotation errors in the widely used yeast mutation collections

   Nature Methods 9:373–378.

   https://doi.org/10.1038/nmeth.1890

   • PubMed
   • Google Scholar
8. 1. Boussadia O
   2. Amiot F
   3. Cases S
   4. Triqueneaux G
   5. Jacquemin-Sablon H
   6. Dautry F

   (1997) Transcription of Unr (upstream of N-ras) down-modulates N-ras expression in vivo

   FEBS Letters 420:20–24.

   https://doi.org/10.1016/s0014-5793(97)01479-8

   • PubMed
   • Google Scholar
9. 1. Boyle EI
   2. Weng S
   3. Gollub J
   4. Jin H
   5. Botstein D
   6. Cherry JM
   7. Sherlock G

   (2004) GO::termfinder--open source software for accessing gene ontology information and finding significantly enriched gene ontology terms associated with a list of genes

   Bioinformatics 20:3710–3715.

   https://doi.org/10.1093/bioinformatics/bth456

   • Google Scholar
10. 1. Brar GA
    2. Yassour M
    3. Friedman N
    4. Regev A
    5. Ingolia NT
    6. Weissman JS

    (2012) High-Resolution view of the yeast meiotic program revealed by ribosome profiling

    Science 335:552–557.

    https://doi.org/10.1126/science.1215110

    • PubMed
    • Google Scholar
11. 1. Breslow DK
    2. Cameron DM
    3. Collins SR
    4. Schuldiner M
    5. Stewart-Ornstein J
    6. Newman HW
    7. Braun S
    8. Madhani HD
    9. Krogan NJ
    10. Weissman JS

    (2008) A comprehensive strategy enabling high-resolution functional analysis of the yeast genome

    Nature Methods 5:711–718.

    https://doi.org/10.1038/nmeth.1234

    • PubMed
    • Google Scholar
12. 1. Carroll KL
    2. Pradhan DA
    3. Granek JA
    4. Clarke ND
    5. Corden JL

    (2004) Identification of cis elements directing termination of yeast nonpolyadenylated snoRNA transcripts

    Molecular and Cellular Biology 24:6241–6252.

    https://doi.org/10.1128/MCB.24.14.6241-6252.2004

    • PubMed
    • Google Scholar
13. 1. Chen J
    2. Tresenrider A
    3. Chia M
    4. McSwiggen DT
    5. Spedale G
    6. Jorgensen V
    7. Liao H
    8. van Werven FJ
    9. Ünal E

    (2017) Kinetochore inactivation by expression of a repressive mrna

    eLife 6:e27417.

    https://doi.org/10.7554/eLife.27417

    • PubMed
    • Google Scholar
14. 1. Cheng Z
    2. Otto GM
    3. Powers EN
    4. Keskin A
    5. Mertins P
    6. Carr SA
    7. Jovanovic M
    8. Brar GA

    (2018) Pervasive, coordinated protein-level changes driven by transcript isoform switching during meiosis

    Cell 172:910–923.

    https://doi.org/10.1016/j.cell.2018.01.035

    • PubMed
    • Google Scholar
15. 1. Cheng Z
    2. Mugler CF
    3. Keskin A
    4. Hodapp S
    5. Chan LY-L
    6. Weis K
    7. Mertins P
    8. Regev A
    9. Jovanovic M
    10. Brar GA

    (2019) Small and large ribosomal subunit deficiencies lead to distinct gene expression signatures that reflect cellular growth rate

    Molecular Cell 73:36–47.

    https://doi.org/10.1016/j.molcel.2018.10.032

    • PubMed
    • Google Scholar
16. 1. Chia M
    2. Tresenrider A
    3. Chen J
    4. Spedale G
    5. Jorgensen V
    6. Ünal E
    7. van Werven FJ

    (2017) Transcription of a 5’ extended mRNA isoform directs dynamic chromatin changes and interference of a downstream promoter

    eLife 6:e27420.

    https://doi.org/10.7554/eLife.27420

    • PubMed
    • Google Scholar
17. Book

    1. Chu AM
    2. Davis RW

    (2008) High-throughput creation of a whole-genome collection of yeast knockout strains

    In: Osterman AL, Gerdes SY, editors. Microbial Gene Essentiality: Protocols and Bioinformatics. Humana Press. pp. 205–220.

    https://doi.org/10.1007/978-1-59745-321-9

    • Google Scholar
18. 1. Churchman LS
    2. Weissman JS

    (2011) Nascent transcript sequencing visualizes transcription at nucleotide resolution

    Nature 469:368–373.

    https://doi.org/10.1038/nature09652

    • PubMed
    • Google Scholar
19. 1. Corbin V
    2. Maniatis T

    (1989) Role of transcriptional interference in the Drosophila melanogaster Adh promoter switch

    Nature 337:279–282.

    https://doi.org/10.1038/337279a0

    • PubMed
    • Google Scholar
20. 1. Core LJ
    2. Waterfall JJ
    3. Lis JT

    (2008) Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters

    Science 322:1845–1848.

    https://doi.org/10.1126/science.1162228

    • PubMed
    • Google Scholar
21. 1. Cox J
    2. Mann M

    (2008) MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification

    Nature Biotechnology 26:1367–1372.

    https://doi.org/10.1038/nbt.1511

    • PubMed
    • Google Scholar
22. 1. Curtin JA
    2. Dane AP
    3. Swanson A
    4. Alexander IE
    5. Ginn SL

    (2008) Bidirectional promoter interference between two widely used internal heterologous promoters in a late-generation lentiviral construct

    Gene Therapy 15:384–390.

    https://doi.org/10.1038/sj.gt.3303105

    • PubMed
    • Google Scholar
23. 1. de Hoon MJL
    2. Imoto S
    3. Nolan J
    4. Miyano S

    (2004) Open source clustering software

    Bioinformatics 20:1453–1454.

    https://doi.org/10.1093/bioinformatics/bth078

    • PubMed
    • Google Scholar
24. 1. de la Cruz J
    2. Iost I
    3. Kressler D
    4. Linder P

    (1997) The p20 and ded1 proteins have antagonistic roles in eif4e-dependent translation in Saccharomyces cerevisiae

    PNAS 94:5201–5206.

    https://doi.org/10.1073/pnas.94.10.5201

    • PubMed
    • Google Scholar
25. 1. Egorov AA
    2. Alexandrov AI
    3. Urakov VN
    4. Makeeva DS
    5. Edakin RO
    6. Kushchenko AS
    7. Gladyshev VN
    8. Kulakovskiy IV
    9. Dmitriev SE

    (2021) A standard knockout procedure alters expression of adjacent loci at the translational level

    Nucleic Acids Research 49:11134–11144.

    https://doi.org/10.1093/nar/gkab872

    • PubMed
    • Google Scholar
26. 1. Emerman M
    2. Temin HM

    (1984) Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism

    Cell 39:449–467.

    https://doi.org/10.1016/0092-8674(84)90453-7

    • PubMed
    • Google Scholar
27. 1. Fox MJ
    2. Gao H
    3. Smith-Kinnaman WR
    4. Liu Y
    5. Mosley AL

    (2015) The exosome component rrp6 is required for RNA polymerase II termination at specific targets of the nrd1-nab3 pathway

    PLOS Genetics 11:e1004999.

    https://doi.org/10.1371/journal.pgen.1004999

    • PubMed
    • Google Scholar
28. 1. Gao Y
    2. Bai X
    3. Zhang D
    4. Han C
    5. Yuan J
    6. Liu W
    7. Cao X
    8. Chen Z
    9. Shangguan F
    10. Zhu Z
    11. Gao F
    12. Qin Y

    (2016) Mammalian elongation factor 4 regulates mitochondrial translation essential for spermatogenesis

    Nature Structural & Molecular Biology 23:441–449.

    https://doi.org/10.1038/nsmb.3206

    • PubMed
    • Google Scholar
29. 1. Ghaemmaghami S
    2. Huh W-K
    3. Bower K
    4. Howson RW
    5. Belle A
    6. Dephoure N
    7. O’Shea EK
    8. Weissman JS

    (2003) Global analysis of protein expression in yeast

    Nature 425:737–741.

    https://doi.org/10.1038/nature02046

    • PubMed
    • Google Scholar
30. 1. Giaever G
    2. Chu AM
    3. Ni L
    4. Connelly C
    5. Riles L
    6. Véronneau S
    7. Dow S
    8. Lucau-Danila A
    9. Anderson K
    10. André B
    11. Arkin AP
    12. Astromoff A
    13. El-Bakkoury M
    14. Bangham R
    15. Benito R
    16. Brachat S
    17. Campanaro S
    18. Curtiss M
    19. Davis K
    20. Deutschbauer A
    21. Entian K-D
    22. Flaherty P
    23. Foury F
    24. Garfinkel DJ
    25. Gerstein M
    26. Gotte D
    27. Güldener U
    28. Hegemann JH
    29. Hempel S
    30. Herman Z
    31. Jaramillo DF
    32. Kelly DE
    33. Kelly SL
    34. Kötter P
    35. LaBonte D
    36. Lamb DC
    37. Lan N
    38. Liang H
    39. Liao H
    40. Liu L
    41. Luo C
    42. Lussier M
    43. Mao R
    44. Menard P
    45. Ooi SL
    46. Revuelta JL
    47. Roberts CJ
    48. Rose M
    49. Ross-Macdonald P
    50. Scherens B
    51. Schimmack G
    52. Shafer B
    53. Shoemaker DD
    54. Sookhai-Mahadeo S
    55. Storms RK
    56. Strathern JN
    57. Valle G
    58. Voet M
    59. Volckaert G
    60. Wang C
    61. Ward TR
    62. Wilhelmy J
    63. Winzeler EA
    64. Yang Y
    65. Yen G
    66. Youngman E
    67. Yu K
    68. Bussey H
    69. Boeke JD
    70. Snyder M
    71. Philippsen P
    72. Davis RW
    73. Johnston M

    (2002) Functional profiling of the Saccharomyces cerevisiae genome

    Nature 418:387–391.

    https://doi.org/10.1038/nature00935

    • PubMed
    • Google Scholar
31. 1. Giaever G.
    2. Nislow C

    (2014) The yeast deletion collection: a decade of functional genomics

    Genetics 197:451–465.

    https://doi.org/10.1534/genetics.114.161620

    • PubMed
    • Google Scholar
32. 1. Gidoni D
    2. Kadonaga JT
    3. Barrera-Saldaña H
    4. Takahashi K
    5. Chambon P
    6. Tjian R

    (1985) Bidirectional SV40 transcription mediated by tandem Sp1 binding interactions

    Science 230:511–517.

    https://doi.org/10.1126/science.2996137

    • PubMed
    • Google Scholar
33. 1. Gray M
    2. Piccirillo S
    3. Honigberg SM

    (2005) Two-Step method for constructing unmarked insertions, deletions and allele substitutions in the yeast genome

    FEMS Microbiology Letters 248:31–36.

    https://doi.org/10.1016/j.femsle.2005.05.018

    • PubMed
    • Google Scholar
34. 1. Green-Willms NS
    2. Fox TD
    3. Costanzo MC

    (1998) Functional interactions between yeast mitochondrial ribosomes and mRNA 5′ untranslated leaders

    Molecular and Cellular Biology 18:1826–1834.

    https://doi.org/10.1128/MCB.18.4.1826

    • PubMed
    • Google Scholar
35. 1. Gudipati RK
    2. Xu Z
    3. Lebreton A
    4. Séraphin B
    5. Steinmetz LM
    6. Jacquier A
    7. Libri D

    (2012) Extensive degradation of RNA precursors by the exosome in wild-type cells

    Molecular Cell 48:409–421.

    https://doi.org/10.1016/j.molcel.2012.08.018

    • PubMed
    • Google Scholar
36. 1. Güldener U
    2. Heck S
    3. Fielder T
    4. Beinhauer J
    5. Hegemann JH

    (1996) A new efficient gene disruption cassette for repeated use in budding yeast

    Nucleic Acids Research 24:2519–2524.

    https://doi.org/10.1093/nar/24.13.2519

    • PubMed
    • Google Scholar
37. 1. Ha T
    2. DiPrima M
    3. Koparde V
    4. Jailwala P
    5. Ohnuki H
    6. Feng J-X
    7. Palangat M
    8. Larson D
    9. Tosato G

    (2022) Antisense transcription from lentiviral gene targeting linked to an integrated stress response in colorectal cancer cells

    Molecular Therapy. Nucleic Acids 28:877–891.

    https://doi.org/10.1016/j.omtn.2022.05.029

    • PubMed
    • Google Scholar
38. 1. Hausler B
    2. Somerville RL

    (1979) Interaction in vivo between strong closely spaced constitutive promoters

    Journal of Molecular Biology 127:353–356.

    https://doi.org/10.1016/0022-2836(79)90335-8

    • PubMed
    • Google Scholar
39. 1. Hinnebusch AG
    2. Ivanov IP
    3. Sonenberg N

    (2016) Translational control by 5’-untranslated regions of eukaryotic mRNAs

    Science 352:1413–1416.

    https://doi.org/10.1126/science.aad9868

    • PubMed
    • Google Scholar
40. 1. Hirschman JE
    2. Durbin KJ
    3. Winston F

    (1988) Genetic evidence for promoter competition in Saccharomyces cerevisiae

    Molecular and Cellular Biology 8:4608–4615.

    https://doi.org/10.1128/mcb.8.11.4608-4615.1988

    • PubMed
    • Google Scholar
41. 1. Hollerer I
    2. Barker JC
    3. Jorgensen V
    4. Tresenrider A
    5. Dugast-Darzacq C
    6. Chan LY
    7. Darzacq X
    8. Tjian R
    9. Ünal E
    10. Brar GA

    (2019) Evidence for an integrated gene repression mechanism based on mrna isoform toggling in human cells

    G3: Genes, Genomes, Genetics 9:1045–1053.

    https://doi.org/10.1534/g3.118.200802

    • PubMed
    • Google Scholar
42. 1. Homann OR
    2. Johnson AD

    (2010) MochiView: versatile software for genome browsing and DNA motif analysis

    BMC Biology 8:49.

    https://doi.org/10.1186/1741-7007-8-49

    • PubMed
    • Google Scholar
43. 1. Hongay CF
    2. Grisafi PL
    3. Galitski T
    4. Fink GR

    (2006) Antisense transcription controls cell fate in Saccharomyces cerevisiae

    Cell 127:735–745.

    https://doi.org/10.1016/j.cell.2006.09.038

    • PubMed
    • Google Scholar
44. 1. Hug N
    2. Longman D
    3. Cáceres JF

    (2016) Mechanism and regulation of the nonsense-mediated decay pathway

    Nucleic Acids Research 44:1483–1495.

    https://doi.org/10.1093/nar/gkw010

    • PubMed
    • Google Scholar
45. 1. Hughes TR
    2. Roberts CJ
    3. Dai H
    4. Jones AR
    5. Meyer MR
    6. Slade D
    7. Burchard J
    8. Dow S
    9. Ward TR
    10. Kidd MJ
    11. Friend SH
    12. Marton MJ

    (2000) Widespread aneuploidy revealed by DNA microarray expression profiling

    Nature Genetics 25:333–337.

    https://doi.org/10.1038/77116

    • PubMed
    • Google Scholar
46. 1. Hughes CS
    2. Moggridge S
    3. Müller T
    4. Sorensen PH
    5. Morin GB
    6. Krijgsveld J

    (2019) Single-pot, solid-phase-enhanced sample preparation for proteomics experiments

    Nature Protocols 14:68–85.

    https://doi.org/10.1038/s41596-018-0082-x

    • PubMed
    • Google Scholar
47. 1. Huh W-K
    2. Falvo JV
    3. Gerke LC
    4. Carroll AS
    5. Howson RW
    6. Weissman JS
    7. O’Shea EK

    (2003) Global analysis of protein localization in budding yeast

    Nature 425:686–691.

    https://doi.org/10.1038/nature02026

    • PubMed
    • Google Scholar
48. 1. Jambhekar A
    2. Amon A

    (2008) Control of meiosis by respiration

    Current Biology 18:969–975.

    https://doi.org/10.1016/j.cub.2008.05.047

    • PubMed
    • Google Scholar
49. 1. Jamieson DJ
    2. Beggs JD

    (1991) A suppressor of yeast spp81/ded1 mutations encodes a very similar putative ATP-dependent RNA helicase

    Molecular Microbiology 5:805–812.

    https://doi.org/10.1111/j.1365-2958.1991.tb00753.x

    • PubMed
    • Google Scholar
50. 1. Jazwinski SM

    (2013) The retrograde response: when mitochondrial quality control is not enough

    Biochimica et Biophysica Acta 1833:400–409.

    https://doi.org/10.1016/j.bbamcr.2012.02.010

    • Google Scholar
51. 1. Jin Y
    2. Eser U
    3. Struhl K
    4. Churchman LS

    (2017) The ground state and evolution of promoter region directionality

    Cell 170:889–898.

    https://doi.org/10.1016/j.cell.2017.07.006

    • PubMed
    • Google Scholar
52. 1. Jinek M
    2. Chylinski K
    3. Fonfara I
    4. Hauer M
    5. Doudna JA
    6. Charpentier E

    (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity

    Science 337:816–821.

    https://doi.org/10.1126/science.1225829

    • PubMed
    • Google Scholar
53. 1. Law GL
    2. Bickel KS
    3. MacKay VL
    4. Morris DR

    (2005) The undertranslated transcriptome reveals widespread translational silencing by alternative 5’ transcript leaders

    Genome Biology 6:R111.

    https://doi.org/10.1186/gb-2005-6-13-r111

    • PubMed
    • Google Scholar
54. 1. Longtine MS
    2. McKenzie A
    3. Demarini DJ
    4. Shah NG
    5. Wach A
    6. Brachat A
    7. Philippsen P
    8. Pringle JR

    (1998) Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae

    Yeast 14:953–961.

    https://doi.org/10.1002/(SICI)1097-0061(199807)14:10<953::AID-YEA293>3.0.CO;2-U

    • PubMed
    • Google Scholar
55. 1. Lorenz MC
    2. Muir RS
    3. Lim E
    4. McElver J
    5. Weber SC
    6. Heitman J

    (1995) Gene disruption with PCR products in Saccharomyces cerevisiae

    Gene 158:113–117.

    https://doi.org/10.1016/0378-1119(95)00144-u

    • PubMed
    • Google Scholar
56. 1. Makeeva DS
    2. Lando AS
    3. Anisimova A
    4. Egorov AA
    5. Logacheva MD
    6. Penin AA
    7. Andreev DE
    8. Sinitcyn PG
    9. Terenin IM
    10. Shatsky IN
    11. Kulakovskiy IV
    12. Dmitriev SE

    (2019) Translatome and transcriptome analysis of TMA20 (MCT-1) and TMA64 (eif2d) knockout yeast strains

    Data in Brief 23:103701.

    https://doi.org/10.1016/j.dib.2019.103701

    • PubMed
    • Google Scholar
57. 1. Malabat C
    2. Feuerbach F
    3. Ma L
    4. Saveanu C
    5. Jacquier A

    (2015) Quality control of transcription start site selection by nonsense-mediated-mRNA decay

    eLife 4:e06722.

    https://doi.org/10.7554/eLife.06722

    • PubMed
    • Google Scholar
58. 1. Marquardt S
    2. Escalante-Chong R
    3. Pho N
    4. Wang J
    5. Churchman LS
    6. Springer M
    7. Buratowski S

    (2014) A chromatin-based mechanism for limiting divergent noncoding transcription

    Cell 157:1712–1723.

    https://doi.org/10.1016/j.cell.2014.04.036

    • PubMed
    • Google Scholar
59. 1. Martens JA
    2. Laprade L
    3. Winston F

    (2004) Intergenic transcription is required to repress the Saccharomyces cerevisiae Ser3 gene

    Nature 429:571–574.

    https://doi.org/10.1038/nature02538

    • PubMed
    • Google Scholar
60. Preprint

    1. Messner CB
    2. Demichev V
    3. Muenzner J
    4. Aulakh S
    5. Röhl A
    6. Herrera-Domínguez L
    7. Egger AS
    8. Kamrad S
    9. Lemke O
    10. Calvani E
    11. Mülleder M
    12. Lilley KS
    13. Kustatscher G
    14. Ralser M

    (2022) The Proteomic Landscape of Genome-Wide Genetic Perturbations

    bioRxiv.

    https://doi.org/10.1101/2022.05.17.492318

    • Google Scholar
61. 1. Moseley JL
    2. Page MD
    3. Alder NP
    4. Eriksson M
    5. Quinn J
    6. Soto F
    7. Theg SM
    8. Hippler M
    9. Merchant S

    (2002) Reciprocal expression of two candidate di-iron enzymes affecting photosystem I and light-harvesting complex accumulation

    The Plant Cell 14:673–688.

    https://doi.org/10.1105/tpc.010420

    • PubMed
    • Google Scholar
62. 1. Neil H
    2. Malabat C
    3. d’Aubenton-Carafa Y
    4. Xu Z
    5. Steinmetz LM
    6. Jacquier A

    (2009) Widespread bidirectional promoters are the major source of cryptic transcripts in yeast

    Nature 457:1038–1042.

    https://doi.org/10.1038/nature07747

    • PubMed
    • Google Scholar
63. 1. Ntini E
    2. Järvelin AI
    3. Bornholdt J
    4. Chen Y
    5. Boyd M
    6. Jørgensen M
    7. Andersson R
    8. Hoof I
    9. Schein A
    10. Andersen PR
    11. Andersen PK
    12. Preker P
    13. Valen E
    14. Zhao X
    15. Pelechano V
    16. Steinmetz LM
    17. Sandelin A
    18. Jensen TH

    (2013) Polyadenylation site–induced decay of upstream transcripts enforces promoter directionality

    Nature Structural & Molecular Biology 20:923–928.

    https://doi.org/10.1038/nsmb.2640

    • Google Scholar
64. 1. Padmore R
    2. Cao L
    3. Kleckner N

    (1991) Temporal comparison of recombination and synaptonemal complex formation during meiosis in S. cerevisiae

    Cell 66:1239–1256.

    https://doi.org/10.1016/0092-8674(91)90046-2

    • PubMed
    • Google Scholar
65. 1. Pelechano V
    2. Wei W
    3. Steinmetz LM

    (2013) Extensive transcriptional heterogeneity revealed by isoform profiling

    Nature 497:127–131.

    https://doi.org/10.1038/nature12121

    • PubMed
    • Google Scholar
66. 1. Porrua O
    2. Libri D

    (2015) Transcription termination and the control of the transcriptome: why, where and how to stop

    Nature Reviews. Molecular Cell Biology 16:190–202.

    https://doi.org/10.1038/nrm3943

    • PubMed
    • Google Scholar
67. Book

    1. Powers EN
    2. Brar GA

    (2021) Performing ribosome profiling to assess translation in vegetative and meiotic yeast cells

    In: Labunskyy V, editors. Ribosome Profiling. Springer. pp. 89–125.

    https://doi.org/10.1007/978-1-0716-1150-0

    • Google Scholar
68. 1. Preker P
    2. Nielsen J
    3. Kammler S
    4. Lykke-Andersen S
    5. Christensen MS
    6. Mapendano CK
    7. Schierup MH
    8. Jensen TH

    (2008) Rna exosome depletion reveals transcription upstream of active human promoters

    Science 322:1851–1854.

    https://doi.org/10.1126/science.1164096

    • PubMed
    • Google Scholar
69. 1. Proudfoot NJ

    (1986) Transcriptional interference and termination between duplicated alpha-globin gene constructs suggests a novel mechanism for gene regulation

    Nature 322:562–565.

    https://doi.org/10.1038/322562a0

    • PubMed
    • Google Scholar
70. 1. Raught B
    2. Gingras AC
    3. Sonenberg N

    (2001) The target of rapamycin (TOR) proteins

    PNAS 98:7037–7044.

    https://doi.org/10.1073/pnas.121145898

    • PubMed
    • Google Scholar
71. 1. Rege M
    2. Subramanian V
    3. Zhu C
    4. Hsieh T-HS
    5. Weiner A
    6. Friedman N
    7. Clauder-Münster S
    8. Steinmetz LM
    9. Rando OJ
    10. Boyer LA
    11. Peterson CL

    (2015) Chromatin dynamics and the RNA exosome function in concert to regulate transcriptional homeostasis

    Cell Reports 13:1610–1622.

    https://doi.org/10.1016/j.celrep.2015.10.030

    • PubMed
    • Google Scholar
72. 1. Saldanha AJ

    (2004) Java treeview--extensible visualization of microarray data

    Bioinformatics 20:3246–3248.

    https://doi.org/10.1093/bioinformatics/bth349

    • PubMed
    • Google Scholar
73. 1. Schaughency P
    2. Merran J
    3. Corden JL

    (2014) Genome-Wide mapping of yeast RNA polymerase II termination

    PLOS Genetics 10:e1004632.

    https://doi.org/10.1371/journal.pgen.1004632

    • PubMed
    • Google Scholar
74. 1. Schmidt K
    2. Xu Z
    3. Mathews DH
    4. Butler JS

    (2012) Air proteins control differential TRAMP substrate specificity for nuclear RNA surveillance

    RNA 18:1934–1945.

    https://doi.org/10.1261/rna.033431.112

    • PubMed
    • Google Scholar
75. 1. Schulz D
    2. Schwalb B
    3. Kiesel A
    4. Baejen C
    5. Torkler P
    6. Gagneur J
    7. Soeding J
    8. Cramer P

    (2013) Transcriptome surveillance by selective termination of noncoding RNA synthesis

    Cell 155:1075–1087.

    https://doi.org/10.1016/j.cell.2013.10.024

    • PubMed
    • Google Scholar
76. 1. Sehgal A
    2. Hughes BT
    3. Espenshade PJO

    (2008) Oxygen-Dependent, alternative promoter controls translation of tco1+ in fission yeast

    Nucleic Acids Research 36:2024–2031.

    https://doi.org/10.1093/nar/gkn027

    • PubMed
    • Google Scholar
77. 1. Seila AC
    2. Calabrese JM
    3. Levine SS
    4. Yeo GW
    5. Rahl PB
    6. Flynn RA
    7. Young RA
    8. Sharp PA

    (2008) Divergent transcription from active promoters

    Science 322:1849–1851.

    https://doi.org/10.1126/science.1162253

    • PubMed
    • Google Scholar
78. 1. Sen ND
    2. Gupta N
    3. K Archer S
    4. Preiss T
    5. Lorsch JR
    6. Hinnebusch AG

    (2019) Functional interplay between DEAD-box RNA helicases ded1 and dbp1 in preinitiation complex attachment and scanning on structured mrnas in vivo

    Nucleic Acids Research 47:8785–8806.

    https://doi.org/10.1093/nar/gkz595

    • PubMed
    • Google Scholar
79. 1. Song W
    2. Li J
    3. Liang Q
    4. Marchisio MA

    (2016) Can terminators be used as insulators into yeast synthetic gene circuits?

    Journal of Biological Engineering 10:19.

    https://doi.org/10.1186/s13036-016-0040-5

    • PubMed
    • Google Scholar
80. 1. Steinmetz EJ
    2. Conrad NK
    3. Brow DA
    4. Corden JL

    (2001) Rna-Binding protein Nrd1 directs poly (a) -independent’3'-end formation of RNA polymerase II transcripts

    Nature 413:327–331.

    https://doi.org/10.1038/35095090

    • PubMed
    • Google Scholar
81. 1. Struhl K

    (1985) Nucleotide sequence and transcriptional mapping of the yeast pet56-his3-ded1 gene region

    Nucleic Acids Research 13:8587–8601.

    https://doi.org/10.1093/nar/13.23.8587

    • PubMed
    • Google Scholar
82. 1. Teng X
    2. Dayhoff-Brannigan M
    3. Cheng W-C
    4. Gilbert CE
    5. Sing CN
    6. Diny NL
    7. Wheelan SJ
    8. Dunham MJ
    9. Boeke JD
    10. Pineda FJ
    11. Hardwick JM

    (2013) Genome-Wide consequences of deleting any single gene

    Molecular Cell 52:485–494.

    https://doi.org/10.1016/j.molcel.2013.09.026

    • PubMed
    • Google Scholar
83. 1. Teodorovic S
    2. Walls CD
    3. Elmendorf HG

    (2007) Bidirectional transcription is an inherent feature of Giardia lamblia promoters and contributes to an abundance of sterile antisense transcripts throughout the genome

    Nucleic Acids Research 35:2544–2553.

    https://doi.org/10.1093/nar/gkm105

    • PubMed
    • Google Scholar
84. 1. Topf U
    2. Uszczynska-Ratajczak B
    3. Chacinska A

    (2019) Mitochondrial stress-dependent regulation of cellular protein synthesis

    Journal of Cell Science 132:jcs226258.

    https://doi.org/10.1242/jcs.226258

    • PubMed
    • Google Scholar
85. 1. Van Dalfsen KM
    2. Hodapp S
    3. Keskin A
    4. Otto GM
    5. Berdan CA
    6. Higdon A
    7. Cheunkarndee T
    8. Nomura DK
    9. Jovanovic M
    10. Brar GA

    (2018) Global proteome remodeling during ER stress involves hac1-driven expression of long undecoded transcript isoforms

    Developmental Cell 46:219–235.

    https://doi.org/10.1016/j.devcel.2018.06.016

    • PubMed
    • Google Scholar
86. 1. van Werven FJ
    2. Neuert G
    3. Hendrick N
    4. Lardenois A
    5. Buratowski S
    6. van Oudenaarden A
    7. Primig M
    8. Amon A

    (2012) Transcription of two long noncoding RNAs mediates mating-type control of gametogenesis in budding yeast

    Cell 150:1170–1181.

    https://doi.org/10.1016/j.cell.2012.06.049

    • PubMed
    • Google Scholar
87. 1. Villa T
    2. Barucco M
    3. Martin-Niclos MJ
    4. Jacquier A
    5. Libri D

    (2020) Degradation of non-coding rnas promotes recycling of termination factors at sites of transcription

    Cell Reports 32:107942.

    https://doi.org/10.1016/j.celrep.2020.107942

    • PubMed
    • Google Scholar
88. 1. Wach A
    2. Brachat A
    3. Pöhlmann R
    4. Philippsen P

    (1994) New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae

    Yeast 10:1793–1808.

    https://doi.org/10.1002/yea.320101310

    • PubMed
    • Google Scholar
89. 1. Wach A
    2. Brachat A
    3. Alberti-Segui C
    4. Rebischung C
    5. Philippsen P

    (1997) Heterologous HIS3 marker and GFP reporter modules for PCR-targeting in Saccharomyces cerevisiae

    Yeast 13:1065–1075.

    https://doi.org/10.1002/(SICI)1097-0061(19970915)13:11<1065::AID-YEA159>3.0.CO;2-K

    • PubMed
    • Google Scholar
90. 1. Wang C
    2. Liu Y
    3. DeMario SM
    4. Mandric I
    5. Gonzalez-Figueroa C
    6. Chanfreau GF

    (2020) Rrp6 moonlights in an RNA exosome-independent manner to promote cell survival and gene expression during stress

    Cell Reports 31:107754.

    https://doi.org/10.1016/j.celrep.2020.107754

    • PubMed
    • Google Scholar
91. 1. Wei W
    2. Pelechano V
    3. Järvelin AI
    4. Steinmetz LM

    (2011) Functional consequences of bidirectional promoters

    Trends in Genetics 27:267–276.

    https://doi.org/10.1016/j.tig.2011.04.002

    • PubMed
    • Google Scholar
92. 1. Wethmar K

    (2014) The regulatory potential of upstream open reading frames in eukaryotic gene expression

    Wiley Interdisciplinary Reviews. RNA 5:765–778.

    https://doi.org/10.1002/wrna.1245

    • PubMed
    • Google Scholar
93. 1. Winzeler EA
    2. Shoemaker DD
    3. Astromoff A
    4. Liang H
    5. Anderson K
    6. Andre B
    7. Bangham R
    8. Benito R
    9. Boeke JD
    10. Bussey H
    11. Chu AM
    12. Connelly C
    13. Davis K
    14. Dietrich F
    15. Dow SW
    16. El Bakkoury M
    17. Foury F
    18. Friend SH
    19. Gentalen E
    20. Giaever G
    21. Hegemann JH
    22. Jones T
    23. Laub M
    24. Liao H
    25. Liebundguth N
    26. Lockhart DJ
    27. Lucau-Danila A
    28. Lussier M
    29. M’Rabet N
    30. Menard P
    31. Mittmann M
    32. Pai C
    33. Rebischung C
    34. Revuelta JL
    35. Riles L
    36. Roberts CJ
    37. Ross-MacDonald P
    38. Scherens B
    39. Snyder M
    40. Sookhai-Mahadeo S
    41. Storms RK
    42. Véronneau S
    43. Voet M
    44. Volckaert G
    45. Ward TR
    46. Wysocki R
    47. Yen GS
    48. Yu K
    49. Zimmermann K
    50. Philippsen P
    51. Johnston M
    52. Davis RW

    (1999) Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis

    Science 285:901–906.

    https://doi.org/10.1126/science.285.5429.901

    • PubMed
    • Google Scholar
94. 1. Wyers F
    2. Rougemaille M
    3. Badis G
    4. Rousselle J-C
    5. Dufour M-E
    6. Boulay J
    7. Régnault B
    8. Devaux F
    9. Namane A
    10. Séraphin B
    11. Libri D
    12. Jacquier A

    (2005) Cryptic Pol II transcripts are degraded by a nuclear quality control pathway involving a new poly (a) polymerase

    Cell 121:725–737.

    https://doi.org/10.1016/j.cell.2005.04.030

    • PubMed
    • Google Scholar
95. 1. Xu Z
    2. Wei W
    3. Gagneur J
    4. Perocchi F
    5. Clauder-Münster S
    6. Camblong J
    7. Guffanti E
    8. Stutz F
    9. Huber W
    10. Steinmetz LM

    (2009) Bidirectional promoters generate pervasive transcription in yeast

    Nature 457:1033–1037.

    https://doi.org/10.1038/nature07728

    • PubMed
    • Google Scholar

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting the paper "Use of a ubiquitous gene-editing tool in budding yeast causes off-target repression of neighboring gene protein synthesis" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor. The reviewers have opted to remain anonymous.

Comments to the Authors:

We are sorry to say that, after consultation with the reviewers, we have decided that this work will not be considered further for publication by eLife, because the conclusions that can be drawn from the work are not general enough for the broad readership of eLife. We hope that the reviewers' comments and suggestions will be helpful for submission to another journal.

Reviewer #1 (Recommendations for the authors):

1. Line 82 – It should be noted that the deletion set strains have long been shown to be imperfect, as there are many types of genetic modifications within the collection. The relevant references are:

Widespread aneuploidy revealed by DNA microarray expression profiling.

Hughes TR, Roberts CJ, Dai H, Jones AR, Meyer MR, Slade D, Burchard J, Dow S, Ward TR, Kidd MJ, Friend SH, Marton MJ. Nat Genet. 2000 Jul;25(3):333-7. doi: 10.1038/77116.

Genome-wide consequences of deleting any single gene.

Teng X, Dayhoff-Brannigan M, Cheng WC, Gilbert CE, Sing CN, Diny NL, Wheelan SJ, Dunham MJ, Boeke JD, Pineda FJ, Hardwick JM. Mol Cell. 2013 Nov 21;52(4):485-94. doi: 10.1016/j.molcel.2013.09.026. Epub 2013 Nov 7.

2. Lines 83-85 – Seamless gene replacement has been available in yeast long before CRISPR/Cas9-based editing became available. One very common way is described in this reference:

Two-step method for constructing unmarked insertions, deletions and allele substitutions in the yeast genome. Gray M, Piccirillo S, Honigberg SM. FEMS Microbiol Lett. 2005 Jul 1;248(1):31-6. doi: 10.1016/j.femsle.2005.05.018.

3. Lines 269-271 – Was it experimentally verified that the ectopic copy of DBP2 was expressed at normal levels?

4. For all of the growth curves shown in Figure 2 and its supplement, we would be better able to assess the growth if the Y axis was a log scale since the cells are presumably growing exponentially. In addition, it's possible that the cell size or clumpiness of the mutants is different compared to wild type in a way that might affect the OD600 reading.

5. Figure 2 and supplement – The authors present convincing data that the mutant phenotypes observed in their dbp2 knockouts are not complemented by the integration of an exogenous copy of DBP2. However, in some sense, this is negative data, the failure to rescue the mutant phenotype. There are some additional strains whose analysis would be informative. Most obviously, if the dbp1 deletion phenotypes are really caused by reduced expression of MRP51, then integration of a second copy of MRP51 would be very strongly predicted to rescue the dpb1 deletion mutant phenotypes. Second, one would expect that the dpb1 and mrp51 deletions would have similar phenotypes, depending on how much residual Mrp51 remains in the dpb1 deletion.

6. For the experiment in Figure 2F, were opposite orientations examined for these cassettes? From what is written in lines 400-402 it sounds like integration of the standard resistance cassettes in the opposite orientation would abolish the off-target effect in at least some of these cases. Is that known or predicted?

7. line 461 -"Ablate" implies complete loss of function, which isn't the case for the examples shown. Use of a different word, such as "reduce" or "impair" would be more accurate. The severity of the effect would be better understood by comparison of the dbp1 and mrp51 deletions.

8. In the text the authors give their reconfigured cassette a simple and clear name – KANMX-ins. Therefore, for Figure 4 and its supplement, it would be clearer if the "KANMX-ins" name was used for the labels instead of the more complex name that is used.

The comments above should be addressed by experimental or writing changes. The key experiment that should be included pertains to comment 5 above. That will test whether the integration of an ectopic copy of MRP51 rescues the dpb1 deletion phenotype and whether the dpb1 deletion phenotype mimics that of an mrp51 deletion. Both of these would greatly fortify the conclusion that the dpb1 phenotype is caused by reduced mrp51 expression.

Reviewer #2 (Recommendations for the authors):

Suggestions for the authors for the presentation of their study.

General: if possible, please make clear at the level of individual figures and legends that all presented observations are reproducible. Examples would be if growth curves are n=2, potentially show both or show average and range, and state in figure legend instead of methods; for western blot quantification please indicate what the quantification represents – a single representative of experiment done more than once or an average of multiple experiments? If more than one, please add an error or range to the values. If one but representative, please state explicitly.

1. Line 191. "Together these data indicate that cassette replacement of the DBP1 ORF drives aberrant transcription independent of strain, cassette selection identity, and experimental conditions. Instead, the observed misregulation of MRP51 appears to be a general side-effect of the cassette-insertion."

This statement is too strong. In the experiment shown, two marker cassettes that may or may not be related are tested, apparently in the identical orientation. It is not clear if hygMX4 is the standard name for this marker. Is what is meant hphMX4? hphMX4 and kanMX6 will both use TEF promoter and terminator and in the case here, it appears that both markers were used in the same orientation, which as we learn later is likely critical for the mechanism, though not explicitly tested. This means that conclusion that the effect is general for cassette insertion has not been demonstrated, and potentially would not even be expected (based on the understood mechanism).

2. In Several places "believe" is used. Consider "propose" "argue" "hypothesize" "favor" in place.

3. line 296 "Insertion of every resistance cassette tested lead to a similar growth defect in a non-fermentable carbon source (Figure 2F), a phenotype which was not dependent on Dbp1 (Figure 2B)."

In the manuscript, as currently written, the concept of bidirectional transcription directed from the marker promoter is left to mention later. Since the tested markers share attributes, the bidirectional model is an attractive explanation at this position in the manuscript and a straightforward test of the proposed mechanism would be to use the markers in the opposite orientation. This experiment was not done but could be done. It has the added value of directly demonstrating that the effect is not innate to the marker but to the specific context in use.

4. For figure 2A, an additional panel with a difference map of some kind between the two strains could have value in visually demonstrating that the selected groups of genes are or are not the only major groups affected. There appear to be other clusters also affected and a description of these could add value to the presentation – for example, if it were discussed what is known about cellular phenotypes when mitochondrial translation is perturbed in a manner similar to reduced expression of components and how these data merely recapitulate previous observations or go beyond to describe impacts on the proteome that have not been described.

5. For figure 2f, the lack of complementation is a negative result because it is not demonstrated that DBP1 at the ectopic locus is transcribed and translated to the same extent as at the native locus. This is a minor concern because of the use of insulated KO cassette later but examination of the described CRISPR allele, or reversed kanMX6 or hphMX4 cassettes, or a CRISPR allele removing the DPB1 promoter and TSS, or introduction of a stop codon early in the ORF would have been a nice addition to this experiment.

6. Figure 2- fs 1A. The WT sample presumably has an error in the measurement. Even though mutant and WT are normalized there should be a way to propagate the error to the WT to indicate it is 100% +/- SD.

7. Line 357. "For the last case, in which UPF1 was replaced with the NATMX6 cassette, an abundant aberrant transcript was observed." While it is noted later, it would be appropriate here to note that upf1∆ itself may alter the stability of cryptic transcripts and this could be the reason for the especially strong transcript seen here.

8. Line 370 "Seamless deletion of the DBP1 ORF by this strategy led to the production of a transcript containing only the 5' and 3' UTRs of DBP1. Expression of this mutant transcript led to the aberrant translation of a dubious ORF contained within the DBP1 3' UTR, not translated in wild type (Figure 3—figure supplement 1).

It is a useful point to make that any genetic alteration may have unintended consequences (and this is a nice additional example) but the wording seems to unnecessarily conflate CRISPR-mediated editing with one specific time of deletion (removal of ORF) instead of the many different ways gene function may be rendered null using CRISPR editing. The statement just above states "effects" on "surrounding genes" plural but is not clear outside of the 3 UTR ORF as MRP51 shows a 1.2X effect- is this latter effect significant? If not, what other genes are affected? It does not seem necessary to use this example as justification for the new generation of selectable markers that are introduced in the next section- their value stands on their own, regardless of the extra effects of particular types of CRISPR editing.

9. Figure 3. An alternative framing would be that selectable markers employing the TEF1 promoter can drive or induce transcription divergently from the cassette in multiple contexts.

10. Line 492 "These features included a gene antisense and 5' to the cassette.."

This could be more accurately described as KO cassettes can have two orientations relative to the knocked out genes and adjacent genes are not technically antisense unless they occupy the same sequence on the other strand. Consider "These features include an upstream gene transcribed divergently from that of the orientation of the selectable marker gene in the cassette".

11. One consideration that would greatly improve the value of the new selectable markers created would be a design strategy that diversifies insulator sequences such that cassettes would not share flanking sequences. One issue with the kanMX6/hphMX4/natMX etc cassettes is that they share TEF promoter and terminator. This means that using them in the same strain reduces the efficiency of use by about 10-fold, due to marker exchange instead of integration into the desired target when one cassette is already in use in a strain. Having additional same sequences on each end can exacerbate this. One possibility would be to switch 5' and 3' positions but another would be to deploy at least two additional terminator regions to reduce marker exchange propensity.

12. Line 608 "c terminally" → "C-terminally" or "carboxy-terminally".

13. Line 634 "100 µM" of what?

14. Line 652 Fischer → Fisher.

15. Line 751 "We then cloned 3 additional cassettes using varying selection (including Saccharomyce S. pombe HIS5: complements S.cerevisiae HIS3, TRP1, and NATr)…"

Please make clear in methods if these markers were used to replace kan in kanMX6 or if insulators were added to those original cassettes and cite the source of the original cassettes (plasmids) and their official cassette names here for the relevant markers.

16. Line 756. 5' RACE. The nature of the sequencing here is not described. Numbers of "reads" are described, but is it instead meant numbers of clones with cDNA ends mapping to indicated positions because that is my understanding of how the GeneRacer III TOPO kit is designed to be employed. Additionally, were RACE products attempted to be mapped within the cassette and not found, or was it not examined? Are the new products at MRP51 within or external to the cassette and how close to the junction? It would be useful to know the distance between the cassette TSSs for the resistance marker and the induced divergent transcription to get an idea of if the properties are consistent with divergent transcription as described for single nucleosome-free regions or if instead the TEF one sequence used in these cassettes actually has an additional promoter region attached

[Editors’ note: The authors appealed the original decision. What follows is the authors’ response to the first round of review.]

Reviewer #1 (Recommendations for the authors):

1. Line 82 – It should be noted that the deletion set strains have long been shown to be imperfect, as there are many types of genetic modifications within the collection. The relevant references are:

Widespread aneuploidy revealed by DNA microarray expression profiling.

Hughes TR, Roberts CJ, Dai H, Jones AR, Meyer MR, Slade D, Burchard J, Dow S, Ward TR, Kidd MJ, Friend SH, Marton MJ. Nat Genet. 2000 Jul;25(3):333-7. doi: 10.1038/77116.

Genome-wide consequences of deleting any single gene.

Teng X, Dayhoff-Brannigan M, Cheng WC, Gilbert CE, Sing CN, Diny NL, Wheelan SJ, Dunham MJ, Boeke JD, Pineda FJ, Hardwick JM. Mol Cell. 2013 Nov 21;52(4):485-94. doi: 10.1016/j.molcel.2013.09.026. Epub 2013 Nov 7.

We agree and have included these references in the revised manuscript. These reported issues with the deletion collection have helped researchers better guide their interpretation of deletion collection data, and in some cases correct errors in publications that are based on these artifacts. However, the deletion collection and data derived from it is still useful, and widely used. The off-target effect that we report here (a divergent, repressive transcript driven from pTEF when inserted at a non-endogenous locus) is not currently being widely considered as researchers design and interpret either small-scale or deletion-collection-based experiments.

2. Lines 83-85 – Seamless gene replacement has been available in yeast long before CRISPR/Cas9-based editing became available. One very common way is described in this reference:

Two-step method for constructing unmarked insertions, deletions and allele substitutions in the yeast genome. Gray M, Piccirillo S, Honigberg SM. FEMS Microbiol Lett. 2005 Jul 1;248(1):31-6. doi: 10.1016/j.femsle.2005.05.018.

Great point. We cite this study in the revised manuscript and hope that we have made this point clear.

3. Lines 269-271 – Was it experimentally verified that the ectopic copy of DBP2 was expressed at normal levels?

Yes, we include these data in the revised manuscript in Figure 2—figure supplement 2D.

4. For all of the growth curves shown in Figure 2 and its supplement, we would be better able to assess the growth if the Y axis was a log scale since the cells are presumably growing exponentially. In addition, it's possible that the cell size or clumpiness of the mutants is different compared to wild type in a way that might affect the OD600 reading.

We have now included quantification of doubling time for all growth experiments throughout the manuscript for clarity. We sonicate yeast prior to OD measurements and culture seeding, which reduces clumping. Cell size differences also do not seem to be an issue in this case.

5. Figure 2 and supplement – The authors present convincing data that the mutant phenotypes observed in their dbp2 knockouts are not complemented by the integration of an exogenous copy of DBP2. However, in some sense, this is negative data, the failure to rescue the mutant phenotype. There are some additional strains whose analysis would be informative. Most obviously, if the dbp1 deletion phenotypes are really caused by reduced expression of MRP51, then integration of a second copy of MRP51 would be very strongly predicted to rescue the dpb1 deletion mutant phenotypes. Second, one would expect that the dpb1 and mrp51 deletions would have similar phenotypes, depending on how much residual Mrp51 remains in the dpb1 deletion.

Thank you for these suggestions. In the revised manuscript, we include rescue experiments showing rescue of all observed dbp1∆::kanMX6 phenotypes by an exogenous copy of MRP51 as suggested (Figure 2B-C; Figure 2—figure supplement 2A-C). As far as the second experiment, we are fortunate this was already done by other labs and suggests exactly what the reviewer predicts. Strains completely lacking mrp51 cannot survive under conditions requiring respiratory growth (Stenger et al., Microbial Cell 2020) so the poor respiratory growth seen in our study is consistent with a hypomorphic phenotype based on the partial decrease in Mrp51 levels resulting from translational downregulation of this gene.

6. For the experiment in Figure 2F, were opposite orientations examined for these cassettes? From what is written in lines 400-402 it sounds like integration of the standard resistance cassettes in the opposite orientation would abolish the off-target effect in at least some of these cases. Is that known or predicted?

We agree that this is an interesting experiment and have included it in the revised manuscript (Figure 3G, Figure 3—figure supplement 2A). This indeed removes the respiratory growth defect that results from Mrp51 down-regulation and helps solidify our model.

7. line 461 -"Ablate" implies complete loss of function, which isn't the case for the examples shown. Use of a different word, such as "reduce" or "impair" would be more accurate. The severity of the effect would be better understood by comparison of the dbp1 and mrp51 deletions.

We have incorporated this suggested edit in the revised manuscript.

8. In the text the authors give their reconfigured cassette a simple and clear name – KANMX-ins. Therefore, for Figure 4 and its supplement, it would be clearer if the "KANMX-ins" name was used for the labels instead of the more complex name that is used.

Thank you for the suggestion, we have implemented the shorter name for these constructs and the newly added insulated plasmids into the relevant figures.

The comments above should be addressed by experimental or writing changes. The key experiment that should be included pertains to comment 5 above. That will test whether the integration of an ectopic copy of MRP51 rescues the dpb1 deletion phenotype and whether the dpb1 deletion phenotype mimics that of an mrp51 deletion. Both of these would greatly fortify the conclusion that the dpb1 phenotype is caused by reduced mrp51 expression.

Thank you for these excellent suggestions, we believe that we were able to address all comments, including point 5 above and hope that the reviewer finds the revised manuscript to be suitable for publication at eLife.

Reviewer #2 (Recommendations for the authors):

Suggestions for the authors for the presentation of their study.

General: if possible, please make clear at the level of individual figures and legends that all presented observations are reproducible. Examples would be if growth curves are n=2, potentially show both or show average and range, and state in figure legend instead of methods; for western blot quantification please indicate what the quantification represents – a single representative of experiment done more than once or an average of multiple experiments? If more than one, please add an error or range to the values. If one but representative, please state explicitly.

This has been done throughout the revised manuscript. For growth curve data we chose to calculate doubling time for each strain so that readers can easily compare mutant growth rates directly and incorporated biological replicate data into these plots. For each growth experiment we now show one representative curve in the main figure and provide doubling time for all biological replicates in the supplemental figure panels listed.

1. Line 191. "Together these data indicate that cassette replacement of the DBP1 ORF drives aberrant transcription independent of strain, cassette selection identity, and experimental conditions. Instead, the observed misregulation of MRP51 appears to be a general side-effect of the cassette-insertion."

This statement is too strong. In the experiment shown, two marker cassettes that may or may not be related are tested, apparently in the identical orientation. It is not clear if hygMX4 is the standard name for this marker. Is what is meant hphMX4? hphMX4 and kanMX6 will both use TEF promoter and terminator and in the case here, it appears that both markers were used in the same orientation, which as we learn later is likely critical for the mechanism, though not explicitly tested. This means that conclusion that the effect is general for cassette insertion has not been demonstrated, and potentially would not even be expected (based on the understood mechanism).

Thank you for the suggestions, we have moved this statement to after the data presented in Figure 2D (previously presented in Figure 2F) and attempted to clarify the language to indicate that the effect is general to multiple cassettes though as the reviewer indicates is specific (not general) to cassette orientation. Additionally, we now have placed diagrams showing cassette makeup at all relevant figures such that redundant or unique sequences between cassettes can be easily determined.

We added the reviewer-suggested experiment of a flipped cassette inserted at the DBP1 locus (Figure 3G), which demonstrates as the reviewer predicts, the off-target effect is specific to cassette orientation as predicted by our model. This new experiment helps solidify our proposed mechanism, whereby bidirectional promoter activity stemming from cassette promoters drives mis-regulation of neighboring genes.

We have also corrected the hygromycin cassette nomenclature and appreciate this note.

2. In Several places "believe" is used. Consider "propose" "argue" "hypothesize" "favor" in place.

Thank you for this suggestion, we have replaced these instances in the revised manuscript.

3. line 296 "Insertion of every resistance cassette tested lead to a similar growth defect in a non-fermentable carbon source (Figure 2F), a phenotype which was not dependent on Dbp1 (Figure 2B)."

In the manuscript, as currently written, the concept of bidirectional transcription directed from the marker promoter is left to mention later. Since the tested markers share attributes, the bidirectional model is an attractive explanation at this position in the manuscript and a straightforward test of the proposed mechanism would be to use the markers in the opposite orientation. This experiment was not done but could be done. It has the added value of directly demonstrating that the effect is not innate to the marker but to the specific context in use.

We appreciate this suggestion and have included this experiment into the revised manuscript (Figure 3G), and as mentioned above believe it helps solidify our model. We have also restructured the manuscript, including swapping Figures 3 and 4, to bring this point up earlier and revised the manuscript title and abstract to emphasize the importance of bidirectional transcription earlier.

4. For figure 2A, an additional panel with a difference map of some kind between the two strains could have value in visually demonstrating that the selected groups of genes are or are not the only major groups affected. There appear to be other clusters also affected and a description of these could add value to the presentation – for example, if it were discussed what is known about cellular phenotypes when mitochondrial translation is perturbed in a manner similar to reduced expression of components and how these data merely recapitulate previous observations or go beyond to describe impacts on the proteome that have not been described.

This is an interesting point. The two clusters that we highlight are the ones for which the changes made sense as direct effects of the genetic change that we engineered (ie. it was possible to make a model for the changes based on known functions of either Dbp1 or Mrp51) but we agree with the reviewer that other clusters are interesting, as well. For example, there is an interesting cluster midway down the tree that shows high levels in lower fractions in the wild-type cells and is extremely highly enriched for the mitochondrial ATP synthase complex (suggesting that this is not highly assembled in the cells with low Mrp51 levels) and a group of proteins that are up in the dbp1∆::kanMX6 cells (just above the 54S cluster), which includes proteins that transport proteins into the mitochondria (Tim54, Tom70, Sdh3) and not much else.

The type of data that we collected is unusual in that it is mass spectrometry of polysome fractions specifically, and thus there is not much comparable data in the literature. We think it may indeed be providing new insight into the cellular response to mitochondrial dysfunction.

5. For figure 2f, the lack of complementation is a negative result because it is not demonstrated that DBP1 at the ectopic locus is transcribed and translated to the same extent as at the native locus. This is a minor concern because of the use of insulated KO cassette later but examination of the described CRISPR allele, or reversed kanMX6 or hphMX4 cassettes, or a CRISPR allele removing the DPB1 promoter and TSS, or introduction of a stop codon early in the ORF would have been a nice addition to this experiment.

Yes, this is a great point. We have confirmed that Dbp1 is expressed to the same extent from the ectopic and endogenous locus using western blots and added these quantifications to the revised manuscript (Figure 2—figure supplement 2D). We have also bolstered these experiments by demonstrating that while ectopic Dbp1 expression does not rescue the dbp1∆::kanMX6 phenotypes, ectopic expression of Mrp51 does (Figure 2B-C and Figure 2—figure supplement 2A-C). We appreciate these suggestions and believe this data improve the flow of the revised manuscript.

6. Figure 2- fs 1A. The WT sample presumably has an error in the measurement. Even though mutant and WT are normalized there should be a way to propagate the error to the WT to indicate it is 100% +/- SD.

Yes, thank you for catching this error, we apologize with the manner by which these results were previously presented and have fixed this plot in the revised manuscript.

7. Line 357. "For the last case, in which UPF1 was replaced with the NATMX6 cassette, an abundant aberrant transcript was observed." While it is noted later, it would be appropriate here to note that upf1∆ itself may alter the stability of cryptic transcripts and this could be the reason for the especially strong transcript seen here.

Yes, we find this to be a very interesting point! We initially did mention it earlier in the text, but some found it distracting without additional data and this led to its position in the Discussion rather than the Results.

In the revision process, we performed further analyses of RNA decay mutants, which revealed a role for RRP6-dependent exosome in degrading most cassette-driven divergent transcripts (Figure 5). This degradation likely depends on NNS-dependent termination pathways, which are responsible for terminating many endogenous divergent transcripts. We suggest that the ability of naturally occurring NNS termination sequences to prevent widespread transcription interference from cassette-driven divergent transcription may explain why some loci are susceptible to severe neighboring gene disruption upon cassette insertion and some are not.

It is interesting that cells lacking RRP6 show even higher divergent transcript levels when a selection cassette is inserted at the UPF1 locus than is seen with cassette-replacement of UPF1 alone (Figure 5D). This may indicate that both NMD and exosome-mediated decay can target divergent cassette-driven transcripts in some cases. Given the lack of control here, however (ie. no case in which a cassette is inserted at the UPF1 locus, but Upf1 expression is normal), the data for this locus is inconclusive. Moreover, we were unable to find additional data to support a role for NMD in degrading cassette-driven transcripts among published datasets, so discuss a possible role for NMD in the revised manuscript only briefly, when the data for UPF1 cassette insertion are first introduced.

8. Line 370 "Seamless deletion of the DBP1 ORF by this strategy led to the production of a transcript containing only the 5' and 3' UTRs of DBP1. Expression of this mutant transcript led to the aberrant translation of a dubious ORF contained within the DBP1 3' UTR, not translated in wild type (Figure 3—figure supplement 1).

It is a useful point to make that any genetic alteration may have unintended consequences (and this is a nice additional example) but the wording seems to unnecessarily conflate CRISPR-mediated editing with one specific time of deletion (removal of ORF) instead of the many different ways gene function may be rendered null using CRISPR editing. The statement just above states "effects" on "surrounding genes" plural but is not clear outside of the 3 UTR ORF as MRP51 shows a 1.2X effect- is this latter effect significant? If not, what other genes are affected? It does not seem necessary to use this example as justification for the new generation of selectable markers that are introduced in the next section- their value stands on their own, regardless of the extra effects of particular types of CRISPR editing.

We attempted to clarify this in the revised manuscript, as we agree this is not needed to justify new cassettes. The major point, which we hope we have improved our presentation of, is that small genomic changes can have unintended consequences on translation of coding sequences produced from neighboring sequences, a topic that is now in the discussion. We realize that the fold-change labels on this figure were unclear in their significance. We had labeled MRP51 to demonstrate that it is relatively unchanged in the Cas9 unmarked dbp1∆, but realize this was unclear without further context.

Based on the reviewer’s feedback, we realized that there was a better organization of the data that we have implemented in the revised manuscript and that we think address several of the reviewer’s concerns. We moved the data for CRISPR-edited cells to Figure 5—figure supplement 2. We swapped the former Figures 3 and 4 and included new data for the reversed cassette insertion to the new Figure 3. This shows that bidirectional promoter activity from the cassette is driving the off-target effects. Finally, the new Figure 4 and Figure 5 now shows that this cassette-driven off-target effect occurs more broadly but masked by exosome-mediated degradation in many cases. We believe this organization solves several problems, including making the source of this off-target effect clear earlier in the manuscript.

9. Figure 3. An alternative framing would be that selectable markers employing the TEF1 promoter can drive or induce transcription divergently from the cassette in multiple contexts.

Yes, good point. With the manuscript reorganization, we believe that it is now more clearly stated that pTEF-based cassette insertion can drive divergent transcripts in multiple contexts. In fact, the rrp6 data that are in Figure 5 now suggest that they always do so.

10. Line 492 "These features included a gene antisense and 5' to the cassette.."

This could be more accurately described as KO cassettes can have two orientations relative to the knocked out genes and adjacent genes are not technically antisense unless they occupy the same sequence on the other strand. Consider "These features include an upstream gene transcribed divergently from that of the orientation of the selectable marker gene in the cassette".

It's true that divergent is the correct term to use here and we have fixed this wording to be clearer. We appreciate this comment.

11. One consideration that would greatly improve the value of the new selectable markers created would be a design strategy that diversifies insulator sequences such that cassettes would not share flanking sequences. One issue with the kanMX6/hphMX4/natMX etc cassettes is that they share TEF promoter and terminator. This means that using them in the same strain reduces the efficiency of use by about 10-fold, due to marker exchange instead of integration into the desired target when one cassette is already in use in a strain. Having additional same sequences on each end can exacerbate this. One possibility would be to switch 5' and 3' positions but another would be to deploy at least two additional terminator regions to reduce marker exchange propensity.

We appreciate this suggestion and fully agree. We have made two adjustments to reduce sequence overlap and created 2 additional insulated plasmids. First, we reasoned that the 5’ insulator flanking the cassette promoter is necessary to prevent the promoter driven off-target effects while the 3’ flanking insulator is dispensable because it sits adjacent to the cassette terminator. This was confirmed by swapping the orientation of the kanMX6 cassette integrated into the DBP1 locus which places the TEF terminator adjacent to MRP51 (Figure 3G, Figure 3—figure supplement 2A). Therefore, we removed the 3’ terminator and created a 5’ only tDEG1 insulated version of the kanMX6-ins1 plasmid (creating kanMX6-ins6, Figure 3H). Next, we cloned another 5’ insulated cassette (kanMX6-ins7, Figure 3H) that only shares the primer amplification sites with kanMX6-ins6. This plasmid contains the tCYC1 terminator 5’ of the C. glabrata PGK1 promoter expressing nat^r, and is terminated with the C. glabrata PGK1 terminator. To confirm the efficacy of these new insulated cassettes we integrated them into the DBP1 locus exactly as done with the prior cassettes and tested whether the resulting strain exhibited a growth defect in non-fermentable carbon sources. Integration of both kanMX6-ins6 and kanMX6-ins7 did not induce growth defects in the non-fermentable carbon source glycerol when used to replace DBP1, signifying their lack of disruption of MRP51 (Figure 3-supplement 2B). We will deposit these improved insulated plasmids, as well as our original proposed insulated cassettes, at Addgene for easy access.

12. Line 608 "c terminally" → "C-terminally" or "carboxy-terminally".

Thank you for this correction, we have updated the text in the revised manuscript.

13. Line 634 "100 µM" of what?

Cycloheximide, thank you for noting this error, we noted this in the revised manuscript and appreciate your keen eye.

14. Line 652 Fischer → Fisher.

Thank you, we have now corrected this typo.

15. Line 751 "We then cloned 3 additional cassettes using varying selection (including Saccharomyce S. pombe HIS5: complements S.cerevisiae HIS3, TRP1, and NATr)…"

Please make clear in methods if these markers were used to replace kan in kanMX6 or if insulators were added to those original cassettes and cite the source of the original cassettes (plasmids) and their official cassette names here for the relevant markers.

These additional selection markers were used to replace kan^r in the kanMX6-ins2 plasmid and their description is listed in supplemental table 1 and schematized in Figure 3. We have more clearly listed the plasmid that each resistance marker was amplified from and noted in the text that each of these new cassettes all share the same flanking sequence as that in the kanMX6-ins2 plasmid. We have also cloned an additional insulated cassette with non-shared flanking sequences that uses the promoter and terminator from C. glabrata PGK1 to express the nat^r gene, to further facilitate multiple gene replacements within the same strain and reduce the likelihood of marker exchange. We also provided diagrams clearly showing cassette makeup to clarify shared or unique sequences between cassettes for these new selection cassettes (please see suggestion 11 for more information).

16. Line 756. 5' RACE. The nature of the sequencing here is not described. Numbers of "reads" are described, but is it instead meant numbers of clones with cDNA ends mapping to indicated positions because that is my understanding of how the GeneRacer III TOPO kit is designed to be employed. Additionally, were RACE products attempted to be mapped within the cassette and not found, or was it not examined? Are the new products at MRP51 within or external to the cassette and how close to the junction? It would be useful to know the distance between the cassette TSSs for the resistance marker and the induced divergent transcription to get an idea of if the properties are consistent with divergent transcription as described for single nucleosome-free regions or if instead the TEF one sequence used in these cassettes actually has an additional promoter region attached

Yes, the reviewer is correct, here the number of “reads” listed signifies the number of cDNA clones whose 5’ end of transcript mapped to that position. We have changed the language in the text to more clearly represent this and have added the information requested. None of the 5’ RACE products sequenced mapped to the cassette and all clones analyzed are listed and depicted in this figure, apart from a single clone whose 5’ end of transcript mapped within the MRP51 ORF and likely represented a fragmented transcript not a true 5’ cDNA end. The finding that these cassette driven transcripts do not contain any cassette sequence is consistent with our previous observations from the mRNA sequencing data in Figure 1 and Figure 1—figure supplement 1 where the 5’ extended transcripts for MRP51 did not include any cassette sequence. Instead, the TSS for the divergent cassette driven transcript is ~530bp away from the TSS of the sense transcript that drives the expression of kan^r, and ~110bp away from the cassette-genome junction. Interestingly, the distance between the coding and divergent transcript produced from the kanMX6 TEF (A. gossypii) promoter appears to be highly similar to that observed at the S. cerevisiae TEF1 locus (Figure 5E; Xu et al., Nature 2009).

Because of this similarity, and the fact that we have sequenced the plasmids and confirmed there are no additional promoters, we believe the divergent transcript originates from the same nucleosome free region as the sense transcript.

Furthermore, we see the divergent transcript-driven phenotypes resulting from Mrp51 disruption with insertion of the TRP1 cassette, which shares only the 5’ and 3’ flanking primer amplification sequences (Figure 2D). However, this common primer sequence is still present in the insulated cassettes (and not itself insulated) so it cannot be responsible.

We should have noted (and do so in the revised manuscript), that pTEF1 has bidirectional promoter activity along with most promoters, but the resultant transcript from the endogenous locus is subject to rapid exosome-based degradation and thus not visible in WT mRNA-seq data (Xu et al. Nature 2009).

References:

Almada, A. E., Wu, X., Kriz, A. J., Burge, C. B. & Sharp, P. A. Promoter directionality is controlled by U1 snRNP and polyadenylation signals. Nature 499, 360–363 (2013).

Ben-Shitrit, T. et al. Systematic identification of gene annotation errors in the widely used yeast mutation collections. Nature Methods 9, 373–378 (2012).

Carroll Kristina L., Pradhan Dennis A., Granek Josh A., Clarke Neil D., & Corden Jeffry L. Identification of cis Elements Directing Termination of Yeast Nonpolyadenylated snoRNA Transcripts. Molecular and Cellular Biology 24, 6241–6252 (2004).

Chen J. et al. Kinetochore inactivation by expression of a repressive mRNA. eLife 6, 10.7554/eLife.27417 (2017)

Cheng Z. and Otto G.M. et al., Pervasive, coordinated protein level changes driven by transcript isoform switching during meiosis. Cell 172, 5:910-923 (2018)

Chia M. et al. Transcription of a 5' extended mRNA isoform directs dynamic chromatin changes and interference of a downstream promoter. eLife 6, 10.7554/eLife.27420 (2017)

Churchman, L. S. & Weissman, J. S. Nascent transcript sequencing visualizes transcription at nucleotide resolution. Nature 469, 368–373 (2011).

Core, L. J., Waterfall, J. J. & Lis, J. T. Nascent RNA Sequencing Reveals Widespread Pausing and Divergent Initiation at Human Promoters. Science 322, 1845–1848 (2008).

Costanzo M. et al. The Genetic Landscape of a Cell. Science 327, 5964:425-431 (2010)

Curtin J.A. et al. Bidirectional promoter interference between two widely used internal heterologous promoters in a late-generation lentiviral construct. Gene Therapy 15, 5:384-390 (2008)

Egorov, A. A. et al. A standard knockout procedure alters expression of adjacent loci at the translational level. Nucleic Acids Research 49, 11134–11144 (2021).

Floor S.N. & Doudna J.A. Tunable protein synthesis by transcript isoforms in human cells. eLife 5, 10.7554/eLife.10921 (2016)

Gherman A., Wang R., Avramopoulos D. Orientation, distance, regulation and function of neighbouring genes. Human Genomics 3, 2:143 (2009)

Gidoni D. et al. Bidirectional SV40 Transcription Mediated by Tandem Sp1 Binding Interactions. Science 230, 4725:511-517 (1985)

Hollerer I. et al. Evidence for an Integrated Gene Repression Mechanism based on mRNA Isoform Switching in Human Cells. G3 9, 4:1045-1053 (2019)

Neil, H. et al. Widespread bidirectional promoters are the major source of cryptic transcripts in yeast. Nature 457, 1038–1042 (2009).

Ntini E. et al. Polyadenylation site–induced decay of upstream transcripts enforces promoter directionality. Nature Struc & Molec Bio 20, 8:923-928 (2013)

Preker, P. et al. RNA Exosome Depletion Reveals Transcription Upstream of Active Human Promoters. Science 322, 1851–1854 (2008).

Schulz, D. et al. Transcriptome Surveillance by Selective Termination of Noncoding RNA Synthesis. Cell 155, 1075–1087 (2013).

Seila, A. C. et al. Divergent Transcription from Active Promoters. Science 322, 1849–1851 (2008).

Stenger M. et al. Systematic analysis of nuclear gene function in respiratory growth and expression of the mitochondrial genome in S. cerevisiae. Microbial Cell 7, 9:234-249 (2020)

Teodorovic, S., Walls, C. D. & Elmendorf, H. G. Bidirectional transcription is an inherent feature of Giardia lamblia promoters and contributes to an abundance of sterile antisense transcripts throughout the genome. Nucleic Acids Res 35, 2544–2553 (2007).

Trinklein N. D. et al. An Abundance of Bidirectional Promoters in the Human Genome. Genome Research 14, 1:62-66 (2004)

Van Dalfsen K.M. et al. Global Proteome Remodeling during ER Stress Involves Hac1-Driven Expression of Long Undecoded Transcript Isoforms. Dev Cell 46, 2:219-235 (2018)

Vander Wende H.M. Meiotic resetting of the cellular Sod1 pool is driven by protein aggregation, degradation, and transient LUTI-mediated repression. bioRxiv https://doi.org/10.1101/2022.06.28.498006 (2022)

Xu Z. et al. Bidirectional promoters generate pervasive transcription in yeast. Nature 457, 7232:1033-1037 (2009)

Author details

1. Emily Nicole Powers

   Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Writing – original draft, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4497-7318

2. Charlene Chan

   Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

3. Ella Doron-Mandel

   Department of Biological Sciences, Columbia University, New York, United States

   Contribution

   Data curation, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

4. Lidia Llacsahuanga Allcca

   Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

5. Jenny Kim Kim

   Department of Biological Sciences, Columbia University, New York, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

6. Marko Jovanovic

   Department of Biological Sciences, Columbia University, New York, United States

   Contribution

   Data curation, Methodology, Project administration

   Competing interests

   No competing interests declared

7. Gloria Ann Brar

   1. Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, United States
   2. California Institute for Quantitative Biosciences (QB3), University of California, Berkley, Berkley, United States
   3. Center for Computational Biology, University of California, Berkeley, Berkeley, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   gabrar@berkeley.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8560-9581

• 1,480

  Page views

• 168

  Downloads

• 1

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.