Yeast strains used in this study are listed in Supplementary Table S1.

Cultures were started by inoculating 5 ml of YP-dextrose medium with colonies from a freshly streaked plate and grown at 30°C with gentle shaking overnight. Next morning, cultures were diluted (1:100) to appropriate volume and grown to A₆₀₀ ∼0.4–0.6. Equal number of cells were used for strand-specific RT-PCR, 3C, ChIP or strand-specific TRO assays.

3C experiments were performed as described previously in (El Kaderi et al., 2012). The primers used for 3C analysis are shown in Supplementary Table S2. The restriction enzymes used for chromatin digestion of IMD4 gene were Alu1 and Dra1. Each experiment was performed with at least four independently grown cultures. The P1T1 PCR signals obtained using P1 and T1 primers flanking the promoter and terminator regions respectively were normalized with respect to F1-R1 PCR signal that amplifies an uncut region within the gene as shown in Figures 1D, 2B. All controls described in El Kaderi et al. (2012) were performed in 3C experiments. These included no-formaldehyde-crosslinking control, no-ligation control and primer efficiency control.

ChIP experiments (crosslinking, cell lysis and isolation of chromatin) were performed as described previously (El Kaderi et al., 2009). Different strains were constructed by tagging TFIIB and the Ccl1 subunit of TFIIH as indicated in Supplementary Table S1. Anti-HA-Agarose beads, used to pull down HA-tagged subunits, were obtained from Sigma. IgG-Sepharose beads, purchased from GE Healthcare, were used to pull down TAP-tagged Ccl1 subunit of TFIIH. For ChIP analysis, primers used for ChIP PCR are shown in Supplementary Table S2. Each experiment was repeated with at least four independently grown cultures. Background ChIP signal for each TFIIB and TFIIH experiment was determined using no tag control version of strains. All TFIIB and TFIIH ChIP signals were first normalized with no tag control, and then with input.

Transcription analysis was performed by RT-PCR approach as described previously (El Kaderi et al., 2009). The primers used for RT-PCR analysis are shown in Supplementary Table S2.

The strand-specific “Transcription Run-On” (TRO) assay was performed as described in (Dhoondia et al., 2017). The primers used for making cDNA and PCR are described in Supplementary Table S2.

The data shown in each figure is the result of at least three biological replicates and six technical replicates. The quantification and statistical analysis were performed as described in (El Kaderi et al., 2012). Error bars represent one unit of standard deviation. p-values were calculated by two-tailed student t-test (Student, 1908).

A vast majority of intron-containing genes in budding yeast have an intron in the promoter-proximal position. A small number of yeast genes carry an intron in the middle of the gene, and an even fewer number have them in the vicinity of the terminator region. It has been shown that a promoter-proximal intron enhances transcription of a gene by facilitating the recruitment of general transcription machinery and by enhancing transcription directionality (Kwek et al., 2002; Das et al., 2007; Damgaard et al., 2008; Bieberstein et al., 2012; Agarwal and Ansari, 2016). The effect of a promoter-proximal intron on transcription has been conserved during evolution as it is exhibited by simple eukaryotes like budding yeast as well as the most complex mammalian systems. There is, however, no concrete evidence regarding the transcription enhancement potential of a terminator-proximal intron. YPL109C is one of the few yeast genes that contain a terminator-proximal intron. We deleted the terminator-proximal intron of YPL109C, and measured transcription of the gene in the presence and absence of the intron (Figure 1A). Steady-state RNA analysis revealed that YPL109C mRNA level decreased by about 10-folds in the absence of the intron (Figure 1B). Transcription Run-On (TRO) analysis, which measured the transcription of nascent mRNA, revealed similar results (Figure 1C). These results demonstrate that even a terminator-proximal intron has the potential to enhance the transcription of the gene.

To better understand the role of the position of an intron in a gene on its transcription regulatory potential, it required inserting the same intron sequence at different positions of a gene and measuring transcription of the gene at every position. We have earlier showed that ACT1 intron enhances transcription of INO1 in an activator-independent manner (Moabbi et al., 2012). Furthermore, we demonstrated that transcription activation potential of ACT1 intron was not due to the presence of a promoter or enhancer sequence element within the intron but was strictly dependent on the splicing potential of the intron. We therefore inserted ACT1 intron into five different positions of INO1 gene following the strategy described in Moabbi et al. (2012). INO1 is an intron-less gene, which is induced in the absence of inositol in the growth medium. We have previously demonstrated enhancement of INO1 transcription under non-inducing conditions upon insertion of an intron near the promoter region of the gene (Moabbi et al., 2012). Here we inserted ACT1 intron at four additional positions of INO1 gene as shown in Figure 3A. The intron was inserted in the 1,602 bp long INO1 gene at 100, 500, 800 and 1,400 bp positions in the coding region of the gene, and one 10 bp downstream of the coding region in the 3′ UTR of the gene. At each position of the intron, transcription was monitored by strand-specific TRO approach as described in Medler and Ansari (2015). TRO assay measures nascent mRNA, which is not affected by half-life of mRNA, and accurately reflects transcription of the gene. In the absence of an intron, there is very low level basal transcription of the gene. Insertion of the intron at 100 bp position enhanced nascent mRNA levels by about 43 times over the intronless version (Figure 3B). At the 500 bp position, there was only 20 times stimulation in transcript level by the intron over the intron-less control. Thus, the enhancement effect of an intron at the 500 bp position registered a decline by about 50% compared to the 100 bp position (Figure 3B). Introns inserted at 800 and 1,400 bp position resulted in a mere 6 times and 8 times increase respectively in transcription over the intron-less control. These results are in agreement with the published results from other organisms as they clearly demonstrate that the enhancement effect of the intron on gene expression decreases with the increase in distance of the intron from the promoter region (Callis et al., 1987; Nesic et al., 1993; Rose, 2004). There was, however, a surprise when we inserted the intron at a position 10 bp downstream of the coding region. The 3′ UTR intron enhanced INO1 transcription by about 17 times (Figure 3B). These results suggest that the proximity to the terminator region partially restores the enhancement potential of an intron on gene expression. Yet, the terminator-proximal intron was not as efficient as the promoter-proximal intron in enhancing transcription of INO1. Although the promoter-proximal intron was spliced with slightly lower efficiency, overall introns were spliced with similar efficiency at each of these five positions (Supplementary Figure S1), thereby suggesting that the observed results were not because of the differential splicing efficacy of introns.

The INO1 gene does not contain a natural intron. The results obtained with INO1 therefore need to be substantiated with a gene containing a natural intron. We, therefore, repeated the experiment with IMD4, which is a natural intron-containing gene. We first constructed a strain containing the intron-less version of IMD4 gene as described in Moabbi et al. (2012). We then inserted ACT1 intron at three different positions of the IMD4 gene; one at 461 bp position (promoter-proximal), a second one at 750 bp position within the coding region (middle), and a third one in 3′ UTR at the position 20 bp downstream of the coding region (terminator-proximal) as shown in Figure 4A. Like INO1, we have previously demonstrated that ACT1 intron enhances transcription of IMD4 in a splicing-dependent manner as mutation of either 5′ or 3′ splice sites completely abolished transcription enhancement potential of the intron (Agarwal and Ansari, 2016). Transcription of the gene was then monitored by strand-specific TRO approach as described earlier. The results show very little detectable transcription in the absence of an intron (Figure 4B). Insertion of the intron at 461 position (promoter-proximal) stimulated transcription of the gene by about 10-fold over the intronless version. The 750 intron (middle) did not result in enhancement of transcription over the intron-less control as the TRO signal obtained was similar to that in the absence of an intron. In the presence of a terminator-proximal intron, however, TRO signal registered an increase of almost six-fold over the intron-less control (Figure 4B). The enhancement effect of terminator-proximal intron is statistically significant as indicated by the p-value of 2.6E-04 (Figure 4B). Although the promoter-proximal intron was spliced with slightly lower efficiency, overall introns were spliced with similar efficiency at each of these three positions (Supplementary Figure S2), thereby suggesting that the observed results were not because of the differential splicing behavior of introns. These results clearly demonstrate that an intron regains its transcription enhancement potential with increasing proximity to the terminator region of the gene in budding yeast. A logical conclusion of these results is that the proximity to both the promoter and terminator regions contributes to the enhancement potential of an intron on gene expression.

We have recently demonstrated a novel role of a promoter-proximal intron in enhancing transcription by conferring directionality to the promoter-bound polymerase to transcribe in the sense direction and inhibit transcription in the upstream anti-sense direction (Agarwal and Ansari, 2016). To determine if the transcription regulatory function of a terminator-proximal intron is also due to its effect on transcription directionality, we performed TRO in the promoter-proximal 800 bp window; 400 bp downstream (mRNA) and upstream (uaRNA) of the promoter as shown in Figure 4A. The terminator-proximal intron stimulated transcription in the promoter-proximal sense direction (mRNA) by about six fold (Figure 4B, solid purple bar), while transcription in the upstream anti-sense direction (uaRNA) decreased by about 2.5 fold over the intron-less control (Figure 4C, diagonal purple pattern bar). The promoter-proximal intron had a similar effect on mRNA and uaRNA levels as expected (Figure 4C, solid red and diagonal red pattern bar). The middle intron, which did not enhance transcription of the gene (Figure 4C, solid green bar), also had no significant effect on uaRNA transcription over the intron-less control (Figure 4C, diagonal green pattern bar).

To better understand the role of the position of an intron on promoter directionality, we calculated the directionality index by dividing nascent mRNA levels with nascent uaRNA levels for the intron inserted at different positions of the IMD4 gene. The directionality index in the presence of a promoter-proximal intron increased by about 24 fold (Figure 4D, red bar), and in the presence of a terminator-proximal intron by about 10 fold (Figure 4D, purple bar) over the intron-less control. In contrast, the middle intron did not affect the directionality index over the intron-less version of the gene (Figure 4D, green bar). Thus, a terminator-proximal intron also enhances transcription directionality in a manner similar to that observed by the promoter-proximal intron.

Our published results demonstrated that the intron-dependent promoter directionality was through gene looping, which is the interaction of the promoter and terminator regions of a gene in a transcription-dependent manner (Agarwal and Ansari, 2016; Al Husini et al., 2020). Juxtaposition of the terminator and promoter regions places termination factors in the vicinity of the promoter region leading to termination of uaRNA transcription. We therefore examined if a terminator-proximal intron also confers looped gene architecture. Gene looping was detected by the “Chromosome Conformation Capture” (3C) approach. This procedure converts the promoter-terminator interaction into quantitatively measurable PCR products obtained using the primer pair flanking the promoter and terminator region, as schematically represented in Figure 2A. 3C analysis was performed with the intron inserted at different positions of the IMD4 gene following the protocol described in (El Kaderi et al., 2012). In all 3C experiments, complete digestion of chromatin between promoter and terminator, crosslinking-dependence and ligation-dependence of the results was routinely monitored. All 3C PCR products were verified by sequencing. Using our modified high resolution 3C approach, we have previously demonstrated a specific interaction of promoter and terminator regions of INO1 in a transcription-dependent manner (El Kaderi et al., 2009; Moabbi et al., 2012). As expected, the 3C assay detected a compact gene architecture with the promoter and terminator in close proximity for the construct with a promoter-proximal intron (Figure 2B, red bar). The promoter-terminator proximity measured in terms of P1T1 PCR signal was five fold more in the construct with the promoter-proximal intron compared to that in the absence of an intron. In the construct with the intron in the middle of the gene, gene architecture was comparable to that in the absence of the intron (Figure 2B, green bar). In the presence of the terminator-proximal intron, however, the promoter-terminator proximity was partially restored as P1T1 PCR signal registered a 3.6 fold increase over the intron-less strain (Figure 2B, purple bar). YPL109C gene, which has a terminator-proximal intron, similarly exhibited an intron-dependent change in gene architecture resulting in the promoter coming in proximity of the terminator of the gene (Figure 1D).

In the mammalian system, a promoter proximal intron has been shown to enhance transcription by facilitating the recruitment or stabilization of the recruited preinitiation complex (PIC) on the promoter (Kwek et al., 2002; Damgaard et al., 2008; Bieberstein et al., 2012). Whether the intron plays a similar role in the recruitment/stabilization of PIC in yeast, however, has never been investigated. Therefore, to determine if the intron facilitates assembly/stabilization of PIC in yeast, we monitored the occupancy of two GTFs, TFIIB and TFIIH, on the promoter of IMD4 in the presence and absence of an intron. We monitored the occupancy of GTFs in a strain that has HA-tagged TFIIB and TAP-tagged cyclin subunit Ccl1 of TFIIH. ChIP was performed to look directly at the occupancy of two GTFs at the promoter and within the 300 bp promoter downstream coding region. Our results show that the promoter occupancy of TFIIB was enhanced by about five-fold in the presence of a promoter-proximal intron over the intron-less control (p-value of 0.01) (Figure 5B, upper panel, red bar). Some signal for TFIIB is detected in the 300 bp promoter downstream region, but the signal is 70–80% less compared to that on the promoter. A comparison of p-values shows that the decrease in TFIIB signal in the promoter-proximal coding region relative to the promoter is statistically significant (Figure 5B, upper panel red bars). Similar results were obtained for Ccl1. The Ccl1 ChIP signal registered a ten-fold increase in the presence of the promoter-proximal intron over no-intron control (p-value of 0.01) (Figure 5B, lower panel, red bar). Like TFIIB, Ccl1 signal also registered a 70–80% decline in the 300 bp promoter-downstream coding region (Figure 5B, lower panel, red bars). These results clearly show that, like in mammalian cells, an intron facilitates the assembly or stabilization of PIC in yeast as well.

Since we have shown that the position of an intron in a gene affects promoter directionality, we next examined if the position of the intron also affects its ability to recruit or stabilize binding of PIC components on the promoter region. We therefore checked occupancy of TFIIB and TFIIH subunit Ccl1 in the presence of the intron in the middle region and the terminator-proximal region of IMD4 as shown in Figure 5A. In the presence of a middle intron, TFIIB ChIP-signal in the promoter region increased by about two fold (Figure 5B, upper panel, green bar), while Ccl1 signal registered a nearly three fold increase (Figure 5B, lower panel, green bar) over the intron-less control. Although there was a small increase in the transcription factors occupancy, it was statistically not significant as indicated by the p-values (p-values 0.67 for TFIIB and 0.08 for Ccl1). Thus, unlike the promoter-proximal intron, the middle intron does not have a significant effect on the promoter occupancy of the general transcription factors. These results are in accordance with the effect of middle intron on transcription and directionality.

The terminator-proximal intron, like the promoter-proximal intron, though has a positive impact on PIC occupancy. The TFIIB promoter ChIP signal increased by about six fold in the presence of the terminator-proximal intron over the intron-less version (p-value 0.005) (Figure 5B, upper panel, purple bar). TFIIH promoter ChIP signal also increased by around three fold in the presence of the terminator-proximal intron, but the increase in signal is statistically not significant (p-value 0.22) (Figure 5B, lower panel, purple bar). These results are corroborated by the TFIIB and TFIIH ChIP-signals in the promoter-proximal coding region. TFIIB ChIP signal exhibits a statistically significant decline in the coding region compared to the promoter (Figure 5B, upper panel), while TFIIH signal does not exhibit a statistically significant change in the coding region relative to the promoter (Figure 5B, lower panel).

These results suggest that a promoter-proximal intron does indeed enhance occupancy of transcription factors at the promoter, while a terminator-proximal intron may only help with occupancy of certain factors of the PIC such as TFIIB. This may also explain why the terminator-proximal intron is not as efficient as the promoter-proximal intron in enhancing transcription.

Since a terminator proximal intron has been shown to affect termination in metazoans, and intron-dependent gene looping involves a terminator-intron interaction as well, we next examined if the presence of an intron in general, and that of the terminator-proximal intron in particular affect termination of transcription. We monitored termination by transcription run-on (TRO) assay as described in Dhoondia et al. (2017) for IMD4 gene with the intron inserted at three different positions as shown in Figure 6A. TRO assay detects the presence of transcriptionally active polymerases on a gene as shown in Figure 6B. To determine the role of the position of an intron on termination of transcription, we calculated the readthrough index (RTI) in constructs with the intron at different positions in the gene. RTI was calculated by dividing the TRO signal intensity beyond the 3′ end of the gene with the signal value before the 3′ end within the coding region. A termination defect results in a higher RTI value, while efficient termination yields a low RTI value. In the presence of an intron in the promoter-proximal position and terminator-proximal position, the RTI value registered a 20-fold and 10-fold decrease respectively over the intronless control (Figure 6C, red bar and purple bar). In the presence of an intron in the middle position, however, the RTI value decreased by a mere 1.2 fold over that in the absence of an intron (Figure 6C, green bar). These results suggest that the position of an intron affects termination of transcription. Both the promoter-proximal and terminator-proximal introns result in efficient termination, while intronless and middle intron versions exhibited inefficient termination. In general, TRO readthrough signal beyond the 3′ end of gene is expected to be higher in the termination defective mutants than in wild type cells (Birse et al., 1998). This was, however, not the case in constructs without an intron and with an intron in the middle position. The TRO readthrough signals in the presence of an intron at all three positions as well as in the absence of intron were very similar (Figure 6B). Whether the position of an intron in yeast affects termination, therefore, needs more scrutiny.