An essential component of a C-terminal domain phosphatase that interacts with transcription factor IIF in Saccharomyces cerevisiae

Abstract

One of the essential components of a phosphatase that specifically dephosphorylates the Saccharomyces cerevisiae RNA polymerase II (RPII) large subunit C-terminal domain (CTD) is a novel polypeptide encoded by an essential gene termed FCP1. The Fcp1 protein is localized to the nucleus, and it binds the largest subunit of the yeast general transcription factor IIF (Tfg1). In vitro, transcription factor IIF stimulates phosphatase activity in the presence of Fcp1 and a second complementing fraction. Two distinct regions of Fcp1 are capable of binding to Tfg1, but the C-terminal Tfg1 binding domain is dispensable for activity in vivo and in vitro. Sequence comparison reveals that residues 173-357 of Fcp1 correspond to an amino acid motif present in proteins of unknown function predicted in many organisms.

Figure 1

Reconstitution of CTD phosphatase activity with recombinant yeast Fcp1. (A) Yeast Fcp1 purified from yeast whole cell extracts [yFCP1, Fraction 62, MonoQ Fraction (19)] and recombinant Fcp1 (ryFCP1) produced in E. coli and purified by Ni-NTA-garose chromatography were separated on a SDS/7.5% polyacrylamide gel and stained with Coomassie blue. An arrow indicates the position of ryFcp1 that contains 21 extra amino acids: six histidines and a HA tag. (B) Assay of yFcp1 and ryFcp1 for CTD phosphatase activity. The assay (19) measures removal of ³²P from the largest subunit of RPII. Lane 1, autoradiographic signal from the phosphorylated form of the largest subunit of yeast RPII. Lanes 2 and 3, signal remaining after treatment with either renatured recombinant (≈0.2 μg) or purified yeast (≈0.02 μg) Fcp1, respectively; these amounts of protein were saturating in the assay.

Figure 2

Sequence and structure of yeast Fcp1. (A) Comparison of yeast and human Fcp1. Amino acid sequences of yeast Fcp1 (top line) and human FCP1a (J.A., H. Xiao, G. Pan, G. Dahmus, M.C., S. Zhang, R. G. Roeder, M. Dahmus, and J.G., unpublished work) (bottom line) were aligned using the sequence alignment program bestfit (Wisconsin sequence analysis package, GCG). Identical amino acids are indicated by a vertical line, highly similar and similar amino acids by a colon and a dot, respectively. (B) Regions of highest similarity and approximate locations of functional motifs in yeast and human Fcp1. Shaded areas are those most similar between yeast and human proteins. “FCP1 homology” refers to the region of similarity presented in C. (C) Multiple alignment of amino acid sequences homologous to yeast Fcp1 residues 173–362 identified by blast searches of GenBank. Bold letters: residues identical in all sequences. Shaded letters: regions of similarity. 1. S. cerevisiae FCP1, Z49704; 2. S. cerevisiae ORF, X90564; 3. S. cerevisiae ORF, U39205; 4. S. cerevisiae ORF, U10555; 5. S. cerevisiae ORF, Z73115; 6. Schizosaccharomyces pombe ORF, Z50142; 7. Brassica campestris EST, L46538; 8. Arabidopsis thaliana EST, T44887; 9. Brugia malayi EST, H48204; 10. Toxoplasma gondii EST, W35521; 11. Caenorhabditis elegans EST, U29536; 12. Mus musculus EST, W29399; 13. Homo sapiens EST, H24417; 14. Homo sapiens EST, H66914; 15. Homo sapiens EST, H95720; 16. Homo sapiens EST, H84869; 17. Homo sapiens FCP1.

Figure 3

Fcp1 is localized to the nucleus. Cells from asynchronous cultures of strain JA830 (fcp1 ΔLEU2 pHA∷FCP1), which contains pJA794 producing HA-Fcp1 from the FCP1 promoter, and W303 (no HA tag) were fixed and stained with anti-HA mAb. Diffuse background staining was detected in non-HA containing W303 (“no tag”); this frame was overexposed intentionally to detect background fluorescence. HA-Fcp1-specific staining colocalized with the nucleus as revealed by 4′,6-diamidino-2-phenylindole staining (HA∷FCP1 panels). Both the 4′,6-diamidino-2-phenylindole and anti-HA images were superimposed onto Nomarski images of the stained cells.

Figure 4

Binding between Fcp1 and RAP74. (A) Various ³⁵S-labeled portions of yeast Fcp1 and human FCP1a, as indicated, were synthesized in rabbit reticulocyte lysates (input), applied to GST and GST-yRAP74 (649–735) microaffinity columns, and processed as described in Materials and Methods. Volumes corresponding to 5% of the input and 25% of each eluate were loaded on the gel. Only the region of each autoradiogram containing full-length in vitro synthesis products is shown. (B) Various ³⁵S-labeled portions of yeast Fcp1 and human FCP1a, as indicated, were chromatographed on GST and GST-hRAP74 (436–517) microcolumns. The columns were run and the eluates analyzed as described for A.

Figure 5

Yeast FCP1 is essential. (A) Diploid strain JA830 disrupted for one copy of FCP1 was induced to sporulate, and all spores of each tetrad were tested for growth. (B) Diploid JA830 was transformed with pJA754 to produce HA-tagged Fcp1 from the GAL10 promoter. Sporulation was induced, and spores were tested for growth on medium containing galactose. (C) Deletion constructs of Fcp1 were tested for complementation of a strain disrupted for FCP1 in vivo and for phosphatase activity in vitro. Amino acids in each construct and locations of RAP74 interacting regions are indicated. Low copy refers to cells heterozygous for the FCP1 disruption transformed with plasmids containing FCP1 under the control of the endogenous promoter (pJA794, pDB3, pJA796, pDB4, and pJA804); cells were sporulated and growth of haploids was tested on selective media. High copy refers to cells heterozygous for the FCP1 disruption transformed with plasmids containing FCP1 under control of the GAL10 promoter (pJA739, pJA756, and pJA798); cells were sporulated and growth of haploids was tested on medium containing galactose. (+) refers to complementation for viability. The in vitro reactions (19) were carried out with Fcp1 proteins overproduced in E. coli and purified as described in Experimental Procedures; (+) indicates that phosphatase activity was detected. nd, not done.