Export of a lariat intermediate requires the mRNA export factor Mex67p.

A. Schematic representation of the brA and brG export reporters. B. The brG reporter accumulates lariat intermediate, as well as pre-mRNA to a lesser extent, as shown by extension of a primer that anneals to the downstream exon. The migration of the indicated splicing species is shown. C. By smRNA FISH, the export of the brG reporter is impeded by the mex67-5 mutant at the non-permissive temperature. Wild-type or mex67-5 mutant cells transformed with the indicated reporters were shifted from 30 °C to 37 °C for 30 minutes. The fraction of nuclear RNA for each cell was quantitated and displayed as a box plot. The number of cells used for quantitation is indicated beneath each box plot. The lacZ region of the reporter transcript was targeted by Cy3-labeled probes; DNA was probed by DAPI. Representative cells are shown; dashed lines mark the cell boundary. D. By RNA co-IP, Mex67p interacts with lariat intermediates in vivo. Top, the indicated cells were mixed before lysis and co-IP. Bottom, the brG lariat intermediate or RPL21a mRNA was detected by RT-PCR before (Input) and after (Bound) IP of Mex67p-TAP, expressed from the indicated strains (MEX67-TAP). The p-values were calculated by Mann-Whitney test; n.s. (not significant), p > 0.05; *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001; ****, p ≤ 0.0001.

Export of a lariat intermediate requires the mRNA export adapters Yra1p, Nab2p, and Npl3p.

A, B. By smRNA FISH, export of the brG reporter requires YRA1. Wild-type or yra1-8-GFP mutant cells were shifted from 25 °C to 37 °C for 2 hours. Panel A shows representative images. Panel B shows the quantitation of the nuclear fraction of reporter RNA. C. By RNA co-IP, Yra1p interacts directly with lariat intermediates. After formaldehyde crosslinking, HA-Yra1p was immunoprecipitated from cell extracts under denaturing conditions with or without antibodies, and then associated brG lariat intermediate or RPL21a mRNA was detected by RT-qPCR. The levels of immunoprecipitated RNA are shown as a percentage of input for three technical replicates; the height of the bar indicates the mean. D, E. By smRNA FISH, the export of the brG reporter requires NAB2. Wild-type or nab2-ΔN mutant cells were shifted from 30 °C to 16 °C for 2 hours. Panel D shows representative images; panel E shows quantitation of the nuclear fraction of reporter RNA. F. By RNA co-IP, Nab2p interacts directly with lariat intermediates. HTB-tagged NAB2-HTB cells and untagged control cells were treated with and without UV, Nab2p-HTB was pulled down from cell extracts under denaturing conditions using Ni-NTA agarose, and then associated brG lariat intermediate or RPL21a mRNA was assayed by qRT-PCR. The levels of pulled-down RNA were quantitated and illustrated as in panel C. G, H. By smRNA FISH, the export of the brG reporter requires NPL3. Wild-type or npl3-1 mutant cells were shifted from 25 °C to 37 °C for 2 hours. Panel G shows representative images; panel H shows the quantitation of the nuclear fraction of reporter RNA. Throughout, for smRNA FISH cells were probed and demarked as in Fig. 1C (top panel); the nuclear fraction was calculated and displayed as in Fig. 1C (bottom panel).

The export of lariat intermediates requires Mlp1p.

A. Schematic representation of the ACT1-IRES-lacZ export reporters. B, C. Efficient export of the brG but not the brA reporter requires MLP1. In panel B, MLP1 or mlp1Δ cells were grown at 30 °C, and the cytoplasmic localization of the reporters was assayed by measuring the β-galactosidase activity of cell extracts. The normalized β-galactosidase activities of four technical replicates are shown; the height of the bar and the number indicate the mean activity, normalized to the mean activity of the same reporter in the MLP1 strain. In panel C, MLP1 or mlp1Δ cells were grown at 30 °C, and the cytoplasmic localization of the reporters was assayed directly by smRNA FISH (left); the nuclear fraction of reporter RNA was quantitated (right). D, E. The export of the UAc lariat intermediate is comprised by the mlp1Δ mutation. In panel D, MLP1, mlp1Δ, dbr1Δ, or mlp1Δ dbr1Δ cells were grown at 30 °C, and the cytoplasmic localization of the brA- and UAc-IRES reporters was assayed by measuring β-galactosidase activity of cell extracts. The β-galactosidase activity was quantitated as in B; values were normalized to the brA-IRES reporter in the wild-type strain (MLP1). In panel E, mutant dbr1Δ or double mutant mlp1Δ dbr1Δ cells were grown at 30 °C and then assayed for reporter localization by smRNA FISH (left); the nuclear fraction of reporter RNA was quantitated (right). Throughout, for smRNA FISH, cells were probed and demarked as in Fig. 1C; the nuclear fraction was calculated and displayed as in Fig. 1C. The p-values were calculated by Mann-Whitney test and represented by asterisks as in Fig. 1.

The export of lariat intermediates requires an interaction between Nab2p and Mlp1p.

A. A deletion in the Nab2p-interacting domain of Mlp1p reduces the export of the brG reporter. Mutant mlp1Δ cells were transformed with a vector control, MLP1, or mlp1Δ1586-1768, grown at 30 °C, and then lysed to assay β-galactosidase activity of the brG reporter. The activity was quantitated as in Fig. 3B; values were normalized to the brG reporter in the MLP1 strain transformed with a vector control. Panels B and C show that a double mutation in the Mlp1p-interacting domain of Nab2p reduces the export of the brG reporter, specifically. In panel B, NAB2 or nab2-F72AF73A cells were grown at 30 °C and then lysed to assay β-galactosidase activity. The activity was quantitated as in Fig. 3B; values for each reporter were normalized to the mean activity of the same reporter in the NAB2 strain. In panel C, NAB2 or nab2-F72AF73A cells were grown at 30 °C and then assayed for reporter localization by smRNA FISH (left); cells were probed and demarked as in Fig. 1C. The nuclear fraction of reporter RNA was quantitated and displayed, on the right, as in Fig. 1C.

The export of lariat intermediate requires Tom1p-mediated ubiquitylation of Yra1p.

A, B. The efficient export of lariat intermediates requires TOM1 (A) and sites of TOM1-dependent ubiquitylation in YRA1 (B). TOM1 and tom1Δ cells (A) were grown at a permissive temperature of 30 °C, and YRA1 and yra1-KR-all cells (B) were grown at semi-permissive temperature of 25 °C. The cytoplasmic localization of the reporters was assayed by measuring β-galactosidase activity of cell extracts. The activity of three technical replicates was quantitated and illustrated as in Fig. 3B; values for each reporter were normalized to the mean activity of the same reporter in the TOM1 strain (A) or YRA1 stain (B). C. The release of Yra1p from lariat intermediate, as for mRNA, requires sites of TOM1-dependent ubiquitylation. Crosslinking and RNA co-IP were executed as in Fig. 2C, except that both YRA1-HA and yra1-KR-all-HA cells were shifted to 37 °C for 2 hours before crosslinking; the levels of Immunoprecipitated RNA were quantitated as in Fig. 2C. D, E. Export of lariat intermediate, as well as mRNA, requires Tom1p (D) and ubiquitylation of Yra1p (E). Compromising the activities of these proteins accumulates lariat intermediate and mRNA in the nucleus. Cells were shifted to the non-permissive temperature of 37 °C for 2 hours and then assayed for reporter localization by smRNA FISH (left); the nuclear fraction of reporter RNA was quantitated (right). F. The nuclear basket factors MLP1 and/or MLP2 are required for the nuclear localization of mRNA and lariat intermediate. Cells were shifted to the non-permissive temperature of 37 °C for 2 hours and then assayed for reporter localization by smRNA FISH (left); the nuclear fraction of reporter RNA was quantitated (right). Throughout, for smRNA FISH cells were probed and demarked as in Fig. 1C; the nuclear fraction was calculated and displayed as in Fig. 1C.

Export of a lariat intermediate requires the mRNA export factor Mex67p but not the tRNA export factor Los1; related to Fig. 1.

A. Schematic representation of the brA and brG export reporters driven by pGPD promoter. B. By RNA-FISH, the export of the brG reporter, like poly(A)+ RNA, is impeded by the mex67-5 mutant at the non-permissive temperature. Cells were shifted from 30 °C to 37 °C for 30 minutes. The intronic region of the brG reporter transcript was detected by Alexa Fluor 488-labelled probes; poly(A)+ RNA was detected in the same cells by a Cy3-labelled poly(dT)50 probe; DNA was probed by DAPI, which is shown separately and overlayed. C. By RNA-FISH, the export of the brG reporter, unlike pre-tRNA, is not impeded by the los1Δ mutant. The intronic region of the brG reporter was detected as in panel A; the intronic region of the tRNALeu precursor was detected in the same cells by a Cy3-labelled probe (note that pre-tRNA splicing does not involve the spliceosome and occurs in the cytoplasm in budding yeast); DNA was probed by DAPI, which is shown separately and overlayed. D. By native RNA co-IP, Mex67-GFP interacts with brG lariat intermediates, which were detected by RT-PCR using lariat specific primers.

Whereas the export of lariat intermediates requires MLP1, the export of pre-mRNA does not; related to Fig. 3.

A. Schematic representation of different ACT1-IRES-lacZ export reporters driven by the pGPD promoter. B, C. The mlp1Δ mutation does not significantly impact the levels of splicing species for the wild-type IRES-containing reporter (brA) or mutated reporters (brG, brC, G1a or G1c), ruling out a role for MLP1 in splicing or RNA stability and consistent with a role for MLP1 in the export of brG lariat intermediates. In panel B, the splicing of each reporter was analyzed in MLP1 and mlp1Δ cells, grown at 30 °C. The levels of splicing precursor, lariat intermediate, and mRNA, were detected by extension of a primer that annealed to the downstream exon; the levels of the snoRNA U14 were detected similarly and served as an internal control; two technical replicates are shown. Panel C shows normalized RNA levels of different splicing species from the two technical replicates in B; the bar height indicates the mean RNA level. Levels of pre-mRNA (salmon), lariat intermediate (yellow), and mRNA (white) were stacked to reflect total 3′ exon levels and normalized to RNA levels of the brA-IRES-lacZ reporter in the MLP1 strain. D. The export of the G1a lariat intermediate is comprised by the mlp1Δ mutation. MLP1 or mlp1Δ cells were grown at 30 °C, and the cytoplasmic localization of the brA- and G1a-IRES-lacZ reporters was assayed by measuring β-galactosidase activity of cell extracts. The activity was quantitated as in Fig. 3B; values were normalized to the brA-IRES-lacZ reporter in the MLP1 strain. E. The export of pre-mRNA is not comprised by the mlp1Δ mutation. MLP1 or mlp1Δ cells were grown at 30 °C, and the cytoplasmic localization of the brA-, G1c- and brC-IRES-lacZ reporters was assayed by measuring β-galactosidase activity of cell extracts. The activity was quantitated as in Fig. 3B; values were normalized to the brA-IRES reporter in the MLP1 strain. The activities for brA reporter were reproduced from D.

The mlp1Δ mutation does not impact the levels of splicing species for either the brA- or UAc-IRES reporters in the presence or absence of Dbr1p; related to Figure 3.

In panel A, MLP1, mlp1Δ, dbr1Δ, or mlp1Δ dbr1Δ cells were grown at 30 °C and splicing of the reporters was analyzed by primer extension, as described in panel Fig. S2B. Panel B shows the quantitation of total RNA levels. Values were normalized to the brA-IRES reporter in the MLP1 strain. Note that the lariat intermediate dominates the splicing species from the UAc-IRES reporter in the dbr1Δ strain, where this species is specifically stabilized.

Deletion of Nab2-interacting domain in Mlp1p does not compromise Mlp1p localization, and nab2-F72A/F73A does not compromise either growth or splicing; related to Figure 4.

A. Mutated mlp1p-Δ1586-1768-GFP localizes to the nuclear periphery, just as wild-type Mlp1p-GFP does. GFP was probed (in live cells) by epi-fluorescence microscopy. Note that the crescent shape is characteristic of Mlp1p localization, which reflects exclusion from the region of the nuclear periphery where the nucleolus resides. Also note that whereas mlp1p-Δ1586-1768-GFP lacks the Nab2p-interacting domain, the mutated protein does include its nuclear localization signal. B. The nab2-F72A/F73A mutant displays no growth defect over a range of temperatures. Growth of the indicated strains was assayed at the indicated temperatures by spotting serial dilutions of cultures onto rich, solid media (YPDA plates) and incubating for 5 days at 16 °C, 4 days at 20 °C, 2 days at 30 °C, and 1 day at 37 °C. C, D. The nab2-F72A/F73A double mutation does not impact the splicing of the brA, G1c, or brG reporters. In panel C, splicing was analyzed by primer extension as in Fig. S2B. Panel D shows the quantitation of total RNA levels. The total RNA of each reporter was normalized to its RNA level in the NAB2 strain.

The yra1-KR-all mutant displayed a mild growth defect at 25 °C; related to Figure 5.

YRA1 and yra1-KR-all cells were streaked onto YPDA, solid media, and incubated for 2 days at 25 °C or 34 °C.

Yeast strains used in this study.

Plasmids used in this study.