Identification of multiple cyclin subunits of human P-TEFb

Abstract

The transition from abortive into productive elongation is proposed to be controlled by a positive transcription elongation factor b (P-TEFb) through phosphorylation of the carboxy-terminal domain (CTD) of the largest subunit of RNA polymerase II. Drosophila P-TEFb was identified recently as a cyclin-dependent kinase (CDK9) paired with a cyclin subunit (cyclin T). We demonstrate here the cloning of multiple cyclin subunits of human P-TEFb (T1 and T2). Cyclin T2 has two forms (T2a and T2b) because of alternative splicing. Both cyclin T1 and T2 are ubiquitously expressed. Immunoprecipitation and immunodepletion experiments carried out on HeLa nuclear extract (HNE) indicated that cyclin T1 and T2 were associated with CDK9 in a mutually exclusive manner and that almost all CDK9 was associated with either cyclin T1 or T2. Recombinant CDK9/cyclin T1, CDK9/cyclin T2a, and CDK9/cyclin T2b produced in Sf9 cells possessed DRB-sensitive kinase activity and functioned in transcription elongation in vitro. Either cyclin T1 or T2 was required to activate CDK9, and the truncation of the carboxyl terminus of the cyclin reduced, but did not eliminate, P-TEFb activity. Cotransfection experiments indicated that all three CDK9/cyclin combinations dramatically activated the CMV promoter.

Figure 1

Sequence comparison of multiple human cyclin T’s with Drosophila cyclin T. (A) Diagram of human cyclins T1, T2a, and T2b and Drosophila cyclin T. Amino acids are numbered on the bottom of each protein. The cyclin box is indicated. Human cyclins T2a and T2b have 642 amino acids in common but different carboxyl termini (black boxes). (B) Sequence alignment of the cyclin boxes. Identity is indicated by reverse shading.

Figure 2

Expression pattern of human P-TEFb subunits. A multiple tissue Northern blot (Clontech) was probed according to the manufacturer. The sequences used for probes are as follows: (CDK9) Total encoding cDNA sequence; (cyclin T1) 0.4 kb encoding the amino terminus; (cyclin T2) 1.35 kb encoding the carboxyl terminus of cyclin T2b; β-actin: 2.0 kb of cDNA. The sizes of molecular markers are at right. (H) Heart; (B) brain; (Pl) placenta; (Lu) lung; (Li) liver; (M) skeletal muscle; (K) kidney; (Pa) pancreas.

Figure 3

Purified recombinant human P-TEFb proteins. (A) Diagram of P-TEFb protein constructs. (a) CDK9/T1; (b) CDK9/T2a; (c) CDK9/T2b; (d) CDK9/T2 (1–286); (e) CDK9. CDK9 is His-tagged. (B) SDS-PAGE (silver stained) of the purified recombinant proteins. (M) 10-kD ladder (10–120 plus 200 kD) with darker 50-kD band.

Figure 4

Multiple cyclin subunits are present and associated with CDK9 in HNE. (A) Western blot IPs. Human P-TEFb was immunoprecipitated using affinity-purified antibodies against GST, CDK9, cyclin T1, or cyclin T2. The IPs (equivalent to 10 μl HNE) were separated on an SDS gel and subjected to Western blotting using antibodies against cyclins T1, T2, or CDK9. The identity of the proteins is indicated at right. Recombinant proteins CDK9/T1, CDK9/T2a, and CDK9/T2b were used as standards. (B) Western blot of immunodepleted HNE. HNE was depleted using antibodies against GST, CDK9, cyclin T1, cyclin T2, or cyclins T1 and T2 together. Depleted HNEs equivalent to 2 μl of HNE (10×) were loaded on an SDS gel followed by Western blotting using antibodies against cyclins T1, T2, CDK9, or p62 (a TFIIH subunit as a loading control). (Right) Original HNE was titrated as a quantitation standard.

Figure 5

Recombinant human P-TEFb proteins have DRB-sensitive CTD kinase activity. (A) CTD kinase assay using Drosophila RNA polymerase II and the indicated amount of recombinant proteins. Labeled RNA polymerase molecules were analyzed on a 6%–15% SDS–polyacrylamide gel by autoradiography. (1×) Approximately 10 fmoles. (IIo and IIa) Shifted and unshifted largest subunit of RNA polymerase II, respectively. (B) DRB inhibition. P-TEFb proteins (5×) were used in the kinase assay with the addition of the indicated amount of DRB. Radiolabeled RNA polymerase II was quantified using a Packard InstantImager and normalized to the starting level (100%).

Figure 5

Recombinant human P-TEFb proteins have DRB-sensitive CTD kinase activity. (A) CTD kinase assay using Drosophila RNA polymerase II and the indicated amount of recombinant proteins. Labeled RNA polymerase molecules were analyzed on a 6%–15% SDS–polyacrylamide gel by autoradiography. (1×) Approximately 10 fmoles. (IIo and IIa) Shifted and unshifted largest subunit of RNA polymerase II, respectively. (B) DRB inhibition. P-TEFb proteins (5×) were used in the kinase assay with the addition of the indicated amount of DRB. Radiolabeled RNA polymerase II was quantified using a Packard InstantImager and normalized to the starting level (100%).

Figure 6

Recombinant human P-TEFb proteins function in transcription. (A,B) Continuous labeling transcription using a CMV promoter was performed with HNE or CDK9-depleted HNE with addition of the indicated amount of recombinant proteins. DRB (50 μm) was added as shown. (C) Plot of runoff. The radioactivity in the runoff transcripts was quantitated using a Packard InstantImager.

Figure 7

Overexpression of human P-TEFb in HeLa cells stimulates transcription from a CMV promoter. Transfections contained 1 million HeLa cells with 0.5 μg of a plasmid with the CMV promoter-driving luciferase as a reporter and 0.25 μg of each plasmid with the CMV promoter-driving CDK9 or the indicated cyclin. Error bars are derived from comparing results from duplicate transfections. This experiment was performed a number of times and the results are typical.