CRISPRi is not strand-specific and redefines the transcriptional landscape

CRISPRi, an adapted CRISPR-Cas9 system, is proposed to act as a strand-specific roadblock to repress transcription in eukaryotic cells using guide RNAs (sgRNAs) to target catalytically inactive Cas9 (dCas9) and offers an alternative to genetic interventions for studying pervasive antisense transcription. Here we successfully use click chemistry to construct DNA templates for sgRNA expression and show, rather than acting simply as a roadblock, binding of sgRNA/dCas9 creates an environment that is permissive for transcription initiation and termination, thus generating novel sense and antisense transcripts. At HMS2 in Saccharomyces cerevisiae, sgRNA/dCas9 targeting to the non-template strand results in antisense transcription termination, premature termination of a proportion of sense transcripts and initiation of a novel antisense transcript downstream of the sgRNA/dCas9 binding site. This redefinition of the transcriptional landscape by CRISPRi demonstrates that it is not strand-specific and highlights the controls and locus understanding required to properly interpret results from CRISPRi interventions.

chemistry is an efficient method for sgRNA construction and that when combined with dCas9, these 47 Here we use the CRISPRi system to study the effect of blocking antisense transcription at loci with 51 well characterised sense:antisense transcript pairs. We have previously used a promoter deletion of 52 the antisense transcript SUT650 at the HMS2 locus to show that SUT650 represses HMS2 sense 53 transcription (Nguyen et al., 2014). Now we use CRISPRi to examine the effects of blocking SUT650 54 antisense transcription to ask (i) whether SUT650 represses HMS2 sense transcription without a 55 genetic intervention and (ii) whether CRISPRi is strand-specific using, in addition to HMS2, an 56 engineered GAL1 gene with a well characterised antisense transcript (Murray et al., 2012(Murray et al., , 2015. 57 The main conclusion from this study is that CRISPRi at the HMS2 locus is not fully strand-specific and 58 results in (i) premature termination of the sense transcript and (ii) initiation of a new unstable 59 antisense transcript in the vicinity of the sgRNA binding site. As transcription from this new antisense 60 initiation site extends in to the HMS2 promoter, there is no net change in HMS2 sense transcript 61 levels. Thus CRISPRi redefines the transcriptional landscape at HMS2. This suggests that routine use 62 of CRISPRi for gene expression analysis will require rigorous analysis of transcript integrity and 63 function before conclusions can be drawn. 64

RESULTS & DISCUSSION 65
DNA templates for sgRNA production can be made using click chemistry 66 CRISPRi-mediated transcriptional repression requires co-expression of a mature sgRNA and dCas9 67 ( Fig 1A). A small library of single-stranded DNAs, comprised of templates for sgRNA variable regions, 68 were joined to the constant region using click chemistry (Fig. 1B) and used successfully as templates 69 for PCR, with no significantly different efficiencies when compared to control full-length synthesised 70 oligonucleotides (Fig. 1C). The PCR products were inserted in place of a URA3 selection cassette in 71 the endogenous snR52 locus for expression of a transcript that is then processed to form the mature 72 nuclear-retained sgRNA (Fig. 1A). Levels of dCas9 protein were uniform between strains ( Fig. 1 -73 Figure Supplement 1). Neither insertion of URA3 into snR52 nor dCas9 expression in the control 74 strains affected growth rate, although strains expressing some sgRNAs grew more slowly indicating a 75 physiological effect (Fig. 1D). 76 CRISPRi represses the production of antisense transcripts at HMS2 and GAL1 77 CRISPRi represses transcription when sgRNAs/dCas9 are targeted to the non-template (NT) strand 78 next to a protospacer adjacent motif (PAM) (Qi et al., 2013) ( Fig. 2A). Firstly, we used an engineered 79 version of GAL1 that has a stable antisense transcript (GAL1 AS) initiating within an ADH1 terminator 80 inserted into the GAL1 coding region (Murray et al., 2012) (Fig. 2B). GAL1 AS is present in cells grown 81 in glucose-containing media when the GAL1 gene is repressed and is reduced as cells are switched 82 into galactose-containing media and GAL1 sense is induced ( Fig. 2C lanes 1-3). We designed sgRNAs 83 adjacent to two PAM sequences on the non-template strand near the antisense transcription start 84 site (TSS) (AS+28NT and AS+112NT) and a third strand-specificity control sgRNA on the template 85 strand in this region (AS+93T) (Fig. 2B). Only sgRNA AS+112NT/dCas9 caused significant (p=0.016) 86 reduction in GAL1 antisense transcript levels, as assessed by Northern blotting (Fig. 2C,D). 87 Next, we examined the HMS2 locus which has a stable antisense transcript, SUT650, initiating within 88 as sequence determinants can also influence repression (Horlbeck, Gilbert, & Villalta, 2016). These 99 results at GAL1 and HMS2 confirm CRISPRi is suitable for reducing levels of antisense transcripts in 100 yeast but controls are needed for each sgRNA designed to ensure that repression has been achieved. 101

CRISPRi repression of SUT650 induces a new shorter antisense transcript at HMS2 102
At HMS2, sgRNA AS+148NT/dCas9 was able to reduce SUT650 levels significantly (Fig. 2F,G). Since 103 SUT650 is a substrate for the major cytoplasmic 5'-3' exonuclease Xrn1 (Fig. 2F, lanes 1&2), we 104 investigated whether the CRISPRi-induced SUT650 reduction was due to direct repression of 105 antisense transcription and/or a reduction in SUT650 stability. Whilst SUT650 was not stabilised 106 upon XRN1 deletion in the strain expressing AS+148NT/dCas9, supporting previous studies that 107 creating an environment that is permissive for transcription initiation (Murray et al., 2012). We note 117 that using probe H4 we were also unable to detect SUT650 initiating from its WT start site but 118 terminating at the AS+148NT binding site, due to its small size (~148 nt). 119

CRISPRi-induced GAL1 AS repression does not affect the GAL1 sense transcript 120
We tested whether GAL1 AS repression by AS+112NT/dCas9 affected the GAL1 sense transcript. 121 Previously we mutated a TATA-like sequence to ablate antisense transcription but observed no 122 difference in GAL1 sense transcript levels in the population (Murray et al., 2015). Using CRISPRi, we 123 also observed no significant change in GAL1 sense transcript levels or size (Fig. 3A,B). To support a strand-specific transcriptional block of AS transcription, GAL1 sense transcript polyA site usage was 125 unaffected by sgRNA AS+112NT/dCas9 (Fig. 3C). Furthermore, XRN1 deletion in this strain also did 126 not affect polyA site usage, ruling out a partial double-stranded transcriptional block and subsequent 127 destabilisation of the resulting truncated sense transcript (Fig. 3C). 128

CRISPRi-induced SUT650 repression truncates the HMS2 sense transcript 129
Previous work shows that reducing SUT650 transcription increases HMS2 sense transcript levels 130 (Nguyen et al., 2014). However, blocking SUT650 by AS+148NT/dCas9 did not similarly increase 131 HMS2 sense levels (Fig. 3D, lanes 1&3). To our surprise, in addition to the full-length HMS2 sense 132 transcript (A), we detected a considerably shortened HMS2 sense transcript (A t ). Since the combined 133 levels of the truncated and full-length transcripts are similar to those in the strain not expressing an 134 sgRNA, independent of the presence of XRN1 (Fig. 3D, lanes 1&3 or 2&6), we hypothesised that the 135 AS+148NT/dCas9 complex bound within the HMS2 coding region may be causing partial premature 136 sense transcription termination, leading to transcript A t . Thus, we mapped the 3' end of HMS2 S by 137 Northern blotting using a series of strand-specific probes across the locus (Fig. 3E). Both transcripts A 138 and A t could clearly be detected using the probes upstream of the AS+148NT/dCas9 binding site 139 (probes H1, H2, H3) but the truncated transcript was undetectable using probe H4 downstream of 140 this site. Thus the transcriptional block caused by AS+148NT/dCas9 is not strand-specific and re-141 defines the transcriptional landscape at the HMS2 locus by creating a chromatin environment that is 142 suitable for both transcription initiation and termination. As an additional control, we designed 143 three sgRNAs to block the HMS2 sense transcripts (S+77NT, S+106NT, S+179NT) and monitored their 144 effects on both HMS2 sense and antisense transcripts (Fig. 3 -Figure Supplement 1). However, we 145 observed no effect on either levels or integrity. 146

CRISPRi-induced antisense repression at HMS2 and GAL1 is not as efficient as previous methods 147
We compared the repression efficiency of CRISPRi with our previously used genetic interventions. At 148 GAL1, mutation of a TATA-like sequence within the ADH1 terminator greatly reduced the level of the GAL1 antisense transcript (Fig. 4A,B) (Murray et al 2015). XRN1 deletion in this strain only slightly 150 increased GAL1 AS levels, suggesting that most repression was at the level of transcription rather 151 than altering transcript stability (Fig. 4B,C). By contrast, XRN1 deletion in the strain expressing 152 AS+112NT/dCas9 did result in some upregulation of GAL1 AS (Fig. 4B,C), indicating that antisense 153 transcript repression was not as great in this strain. Unlike at HMS2 (Fig. 2F,G), this stabilised GAL1 154 AS transcript was the same size as in the control and so likely represents stabilisation of the 155 transcripts escaping repression rather than novel transcripts. 156 Next we compared the efficacy of AS+148NT/dCas9 with previous experiments to ablate SUT650, 157 where we replaced the entire HMS2 coding region, including the antisense TSS with the HMS2-BAT2 di-cistronic transcripts were altered (Fig. 4F). The discovery that whilst AS+148NT/dCas9 163 represses SUT650, a new unstable HMS2 antisense transcript is induced and the sense transcript is 164 prematurely terminated could explain why blocking SUT650 using CRISPRi and the URA3 gene body 165 replacement strategies did not give the same results (Fig. 4F). Thus CRISPRi is not as effective as a 166 genetic mutation in reducing levels of either the GAL1 or HMS2 AS transcripts. 167

Concluding remarks 168
Although CRISPRi has been used to strand-specifically repress antisense transcription at GAL10 169

Yeast growth 298
The strains used in this study are listed in Supplementary File 3. Strains were grown to mid-log at 299 30°C in complete synthetic media lacking leucine (for dCas9 plasmid selection). For experiments 300 studying the GAL1 locus, yeast were grown to mid-log in rich media (YP 2% D/YP 2% Gal) so that the 301 A. An outline of the experiment. Templates for sgRNA production were generated using a click 324 chemistry reaction between a single-stranded DNA oligonucleotide with a 3' alkyne group encoding 325 the constant region (dark blue) and a number of single-stranded DNA oligonucleotides with 5' azido 326 groups encoding the different variable regions (green). The resulting single-stranded DNA 327 oligonucleotide was purified, amplified by PCR and transformed into yeast to replace the URA3 328 selection cassette that had previously been introduced into the endogenous snR52 locus. Correct