| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
* Center for Cardiovascular Development and Department of Medicine;
Departments of Molecular and Cellular Biology, and Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, Texas 77030
 |
ABSTRACT |
|---|
 TOP ABSTRACT I. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
|
I. Overview |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
How does the heart "know" that it should work harder? Ideas regarding some sort of mechanical sensor, which a priori should exist, have been proposed for decades. However, no responsible molecule has been identified conclusively for initiating mechanical signal in the heart. Among the divergent classes of molecules suggested as serving this role are proteins that couple the extraceullar matrix to the cytoskeleton (e.g., integrin-binding proteins, focal adhesion kinase). A muscle LIM protein (MLP)/telethonin (T-cap) complex in the Z-band of the sarcomere may be important in cardiomyocyte stretch sensing (Knoll et al., 2002). One recent study suggests that cell stretch causes a calcium-permeable, growth factor-regulated channel of the transient receptor potential family to translocate to the sarcolemma; overexpression of this channel causes cardiomyopathy in mice (Iwata et al., 2003). For some cells, stretch-activated nuclear calcium channels mediate the transcriptional response to changes in cell shape (Itano et al., 2003). It is not known if this is true for the heart. While the most-proximal stem in mechanotransduction is still the least defined, functionally important downstream signals generated because of load on the heart are many and varied.
As in applying Koch’s postulates of causality to other biological problems, dissecting the cascades for cardiac hypertrophy entails finding mediators that ultimately fulfill the following experimental critieria: 1) they are induced or more active after load; 2) they are sufficient to promote one or more components of hypertrophy when supplied exogenously or expressed as a gain-of-function mutation; and 3) they can be shown to be necessary for one or more components of hypertrophy when their repression or function is blocked. Perhaps the best-posed proximal mediators for cardiac hypertrophy are heterotrimeric G proteins (Adams and Brown, 2001). G protein-coupled receptors of relevance to cardiac hypertrophy include those for the cardiac agonists angiotensin II, endothelin-1 (ET-1), and 
1-adrenergic ligands, each of which has trophic effects on cardiac myocytes apart from peripheral actions such as vasoconstriction. The G proteins implicated in cardiac hypertrophy include Gq, G11, Gs, and Gi, with the case being especially strong for Gq and G11, acting redundantly. Other components of this signal transduction pathway include but are not limited to the mitogen-activated protein kinase (MAPK) family, protein kinase C (PKC), phosphatidylinositol 3-kinase (PI3-K), Akt, calmodulin-dependent protein kinases, and the calmodulin-dependent phosphatase calcineurin (Molkentin et al., 1998; Bueno et al., 2000; Zhang et al., 2000; Chen et al., 2001; Molkentin and Dorn, 2001; Crackower et al., 2002; Minamino et al., 2002; Shioi et al., 2002; Olson and Schneider, 2003). Of these, we will return to Gq and calcineurin subsequently.
A central event for signal transduction in hypertrophy is the modification of ubiquitous and cardiac-enriched transcription factors. The former group includes cyclic AMP (cAMP)-responsive element binding protein/activating transcription factor (CREB/ATF), nuclear factor of activated T cells (NFAT), class II histone deacetylases, and histone acetylases (p300, CREB-binding protein), whereas the latter includes serum response factor (SRF), myocyte enhancer factor-2 (MEF2), homeodomain protein Nkx2.5, and zinc finger protein GATA4, with proteins of both classes serving as coactivators. Transcriptional adaptors that promote the assembly of multifactor complexes are themselves the target of signaling cascades that control key functional properties, such as histone acetylase activity or nuclear localization. Among other targets, these alterations turn on the set of genes known as immediate-early transcription factors (e.g., Fos, Myc, Jun, Egr-1). Certain of the newly synthesized, repressed, and modified proteins influence upstream components of the hypertrophic pathway, activating their activators, activating parallel modules, or providing a negative-feedback loop. Relationships among the pathways for hypertrophy are labyrinthine but no more so than, say, for the cell cycle in proliferating cells. Like "transformation" in cancer biology, which is more-instructively viewed in terms of its discrete elements (e.g., loss of replicative senescence, escape from apoptosis, tumor angiogenesis, metastasis), the hypertrophic phenotype is an aggregate of modular responses — including plasticity of cardiac gene expression (activation of a largely fetal-like gene program), production of growth factors and cytokines, apoptosis, and fibrosis — of which myocyte enlargement is the pathognomic one.
The physical enlargement of cardiomyocytes occurs through a global increase in RNA and protein content per cell. Knowledge concerning the mechanisms for hypertrophic growth is surprisingly scant, compared to the wealth of new insights concerning transcriptional plasticity (Olson and Schneider, 2003). In this review, we consider recent discoveries that implicate the phosphorylation of RNA polymerase II (RNAPII) in cardiac hypertrophy at the sites that are crucial for transcription elongation (Sano et al., 2002; Sano and Schneider, 2003). What kinases are responsible? How do hypertrophic cascades couple to the kinases? What is the evidence that changes in the RNAPII-directed protein kinase circuit are causative with respect to growth? Is such growth adaptive, phenotypically neutral, or adverse?
|
II. The RNAPII Phosphorylation-Dephosphorylation Cycle |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
RNAPII is recruited to promoters in its hypophosphorylated form (RNAPIIa) and subsequently phosphorylated at Ser2 and Ser5 of the heptapeptide repeat, in the transition from initiation to elongation (Komarnitsky et al., 2000; Lin et al., 2002; Orphanides and Reinberg, 2002; Pokholok et al., 2002; Shim et al., 2002; Shilatifard et al., 2003). Importantly, the state of phosphorylation of the CTD is critical to transcription initiation, elongation, and mRNA processing, rather than merely denoting this change of state. Two kinases responsible for phosphorylation of the CTD are cyclin-dependent kinase 7 (Cdk7) (Roy et al., 1994) and Cdk9 (Zhu et al., 1997). Cdk7 phosphorylates Ser5 of the heptapeptide repeat between initiation and promoter clearance, whereas Cdk9 phosphorylates the CTD at Ser2, allowing RNAPII to overcome proximal promoter pausing and migrate into the open reading frame (Orphanides and Reinberg, 2002). The names of these two kinases allude to cyclins as their principal binding partners: cyclin H for Cdk7 (Makela et al., 1994) and cyclin T or K for Cdk9 (Peng et al., 1998a,b). The cyclin H/Cdk7 complex has a third component with the more-than-usually memorable name, ménage-a-trois (MAT1) (Devault et al., 1995; Tassan et al., 1995a). In carrying out its role as a CTD kinase, this trimeric complex is part of the larger general transcription factor TFIIH (Serizawa et al., 1995; Shiekhattar et al., 1995). Cyclin H/Cdk7 also functions as a Cdk-activating kinase (CAK), regulating different phases of the cell cycle by activating other Cdks (Fisher and Morgan, 1994; Harper and Elledge, 1998). The presence of MAT1 in the complex is thought to shift the substrate preference towards the RNAPII CTD (Rossignol et al., 1997; Yankulov and Bentley, 1997).
The cyclin T/Cdk9 heterodimer is known as positive transcription elongation factor-b (P-TEFb) (Marshall and Price, 1992,1995; Conaway and Conaway, 1999; Price, 2000). The cyclin T family is made up of three isoforms — cyclin T1, cyclin T2a, and cyclin T2b — which are known to be functionally distinct in some settings. More distantly related, cyclin K also activates Cdk9. Among the known specificities, the best-studied is Cdk9’s recruitment to the human immunodeficiency virus (HIV) genome exclusively by cyclin T1, via the HIV Tat protein and a 59-base transactivation-responsive region (TAR) in the 5' untranslated region (UTR) of nascent HIV transcripts (Wimmer et al., 1999). Several transcription factors are reported to bind cyclin T2 preferentially (Simone et al., 2002a,b). Negative regulators of transcript elongation include negative elongation factor (NELF) and DRB sensitivity-inducing factor (DSIF) (Wada et al., 2000; Yamaguchi et al., 2002). To inhibit positive transcription elongation, DSIF binds to RNAPII directly and NELF binds to nascent RNA transcripts and RNAPII/DSIF (Yamaguchi et al., 2002). The ability of Cdk9 to promote elongation occurs in part by impairing the association of RNAPII with these key inhibitors. Thus, CTD phosphorylation by Cdk9 enables RNAPII to migrate into the open reading frame and differing elongation factors are bound to the polymerase, depending on position.
Dephosphorylation of the CTD takes place after transcription termination, through the action of a specific TFIIF-associated CTD phosphatase, FCP1 (Cho et al., 1999; Kobor et al., 1999; Kobor and Greenblatt, 2002; Lin et al., 2002; Kamada et al., 2003). FCP1 may preferentially dephosphorylate Ser2 of the CTD (Hausmann and Shuman, 2002) (cf. Lin et al., 2002), whereas a new family of small CTD phosphatases preferentially dephosphorylates the CTD at Ser5 (Yeo et al., 2003). By catalyzing this dephosphorylation after transcription termination, FCP1 prepares RNAPII for another round of transcription. FCP1 likely has an additional role in transcript elongation beyond its function as a CTD phosphatase (Kobor and Greenblatt, 2002; Mandal et al., 2002). Both functions are regulated through its phosphorylation by protein kinases that are yet to be identified (Friedl et al., 2003).
The RNAPII CTD is also the substrate for a third Cdk complex, cyclin C/Cdk8. In this case, however, the kinase acts as a negative regulator of transcription, by phosphorylating the CTD before RNAPII is recruited to promoters (Tassan et al., 1995b; Hengartner et al., 1998; Akoulitchev et al., 2000). Hyperphosphorylated RNAPII cannot be recruited to promoters; thus, Cdk8 prevents transcription from commencing. Cdk8 also phosphorylates cyclin H, repressing the CTD kinase activity of TFIIH and disrupting its ability to activate transcription (Akoulitchev et al., 2000).
This phosphorylation-dephosphorylation cycle — in its relation to the cardiac growth signals we studied — is illustrated, in abbreviated form, in Figure 1 (upper panel).
|
|
III. Cdk7 and Cdk9 Are Targets of the Signaling Cascades for Biomechanical Stress |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
Conversely, both Cdk7 and Cdk9 were activated during chronic cardiac hypertrophy in vivo, provoked by means of biomechanical stress (partial aortic constriction) and by two of its genetically best-proven downstream mediators, Gq (Adams et al., 1998; Wettschureck et al., 2001) and calcineurin (Molkentin et al., 1998; Rothermel et al., 2001). In contrast, acute mechanical stress caused activation only of Cdk9, not Cdk7 (Figure 1a, right). The same was true for isolated neonatal rat cardiomyocytes challenged with ET-1 (Sano et al., 2002). Hence, the principal difference observed was between long-term and short-term growth signals, not between hypertrophy of the intact heart and its surrogate in cultured cells.
The preference for Cdk9 activation as a ubiquitous or at least highly generalizable response to hypertrophic cues prompted the hypothesis that Cdk9 might be the more important of these two CTD kinases in cardiac hypertrophy. Furthermore, Cdk9 has the important technical and conceptual advantages inherent to being an immediate-early response and, hence, is perhaps more likely to be direct. Using catalytically inactive dominant-negative (dn) forms of both kinases, it was demonstrated that dn Cdk9 was sufficient to block ET-1-induced hypertrophy in cultured cardiomyocytes, whereas dn Cdk7 was not (Sano et al., 2002) (Figure 1b). Under these conditions, at least, Cdk9 was essential for myocyte growth. As other critieria to support this interpretation, similar results were obtained using 5,6-dichloro-1-D-ribofuranosylbenzimidazole, a pharmacological inhibitor of transcript elongation that is active preferentially against Cdk9. Furthermore, the phosphorylation of RNAPII triggered by ET-1 occurred at Ser2 of the heptapeptide repeat, the site preferred by Cdk9 (Sano et al., 2002).
|
IV. Trophic Signals Dissociate an Endogenous Inhibitor from Cdk9 |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
q, calcineurin, and biomechanical stress — and was a unitary mechanism possibly shared across this series of interrelating signals? A more-specific mystery was that many common and expected mechanisms were easily excluded. Unlike the easy and intuitively satisfying explanation for downregulation of CTD kinase activity with age (the observed downregulation of all four proteins, the two kinases and their two cyclins), no analogous change was occurring as an early response to trophic signals to account for the rapid reinduction of Cdk9 activity: neither levels of Cdk9, levels of the T cyclins, nuclear localization, nor physical assembly of cyclin T/Cdk9 heterodimers increased (Sano et al., 2002). In 2001, two groups independently identified an astounding endogenous inhibitor of the cyclin T/Cdk9 complex, namely "7SK," a small nuclear RNA (snRNA) of no prior established function (Nguyen et al., 2001; Yang et al., 2001). Based on this breakthrough, the presence of the inhibitor in physical association with cardiac P-TEFb was ascertained directly, using immune complex reverse transcription-polymerase chain reaction (RT-PCR) and sequencing the recovered cDNA (Sano et al., 1999). Binding was confirmed independently using biotinylated RNA complementary to 7SK snRNA for affinity purification, then Western blotting for the co-recovered proteins. In all the hypertrophic models we studied, acute and chronic alike, the activation of Cdk9 occurred through dissociation of this 7SK transcript (Sano et al., 2002) (Figure 1c).
The next question asked was, if signals for cardiac hypertrophy can activate Cdk9, can activating Cdk9 suffice to cause hypertrophy? As pivotal evidence supporting a causal role of the endogenous inhibitor, antisense 7SK was sufficient to increase Cdk9 activity and RNA production (Sano et al., 2002) (Figure 1c). Thus, pharmacological, physiological, and genetic instigators of hypertrophy each removed an inhibitor from Cdk9, whereas removing the inhibitor, by suppressing its expression, was enough to activate Cdk9 and elicit spontaneous growth. Whereas this antisense study ablating 7SK snRNA answered the question conclusively for cultured cells, a corresponding answer for mouse myocardium was sought by forcibly expressing cyclin T1 at the level normal for embryonic or neonatal myocardium. To accomplish this, transgenic mice were engineered using the cardiomyocyte-specific -myosin heavy chain (MHC) promoter to overexpress cyclin T1 in the heart. Cdk9 activation and concentric cardiac hypertrophy resulted, in a dose-dependent fashion, evidenced by an increase in the heart-weight-to-body-weight ratio and in myocyte size (Sano et al., 2002) (Figure 1d).
|
V. Cdk9 as a Therapeutic Target |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
First, if Cdk9 is to be a therapeutic target in humans and not just in mice, then it must be demonstrably germane to human heart disease. In preliminary studies of human heart failure, we recently confirmed the predicted increase in Cdk9 and Cdk7 activity (M. Sano and M.D. Schneider, unpublished results).
Second, even if this is the case, does increasing Cdk9 activity confer increased risk of developing heart failure? Cyclin T1 transgenic mice, our model of Cdk9 activation, appear grossly normal, apart from heart size, under unstressed conditions. Increasingly, however, there is recognition that manipulations of the genome can be incompletely informative or even misleading, if the "phenome" is studied merely as a basal state. This was true, for example, in our prior studies of a proximal MAPK kinase kinase, MEKK1. With little or no findings from the germline deletion by itself, a critical role for MEKK1 was unmasked when tested in the context of a Gq gain-of-function: deleting MEKK1 prevented Gq-induced growth (Minamino et al., 2002). Like Gq at the dose studied, cyclin T1 was well tolerated as a gain-of-function mutation, expressed at the level normal in earlier stages of life, each transgene resulting separately in mild concentric hypertrophy without cell death, fibrosis, or overt dysfunction. However, crossing cyclin T1 with Gq transgenic mice — creating the double-transgenic model — increased Cdk9 activity more so than either alone, increased the heart-weight-to-body-weight ratio more than either of the single transgenics, and induced apoptosis, with the functional sequelae of heart failure and early demise (M. Sano and M.D. Schneider, unpublished data). Completely blocking the increase of Cdk9 function in cardiomyocytes with dnCdk9 prevented Gq-induced growth, yet was tolerated poorly under stress. The same exacerbation and mortality were true, using mechanical load as the hypertrophic signal, in concert with dnCdk9 or cyclin T1. This suggests that Cdk9 is an essential protein at some intermediate level of activity and that graded, titratable inhibition, whether genetic or drug based, would more likely be of benefit than complete disruption of this signaling module.
|
VI. Unanswered Questions and Future Prospects |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
q and the calcium-dependent phosphatase calcineurin. Each of these in transgenic mice reproduces and may be essential for the hypertrophic phenotype (Sano et al., 2002). In acute load or cell-culture models of cardiac hypertrophy, only Cdk9 activity increased. Across this gamut of hypertrophic triggers (hemodynamic, genetic, and pharmacological), a unifying property was activation of Cdk9 not by increased assembly with its functional partner cyclin T but, rather, through dissociation of a remarkable endogenous inhibitor, the noncoding 7SK snRNA. Cdk9 function was sufficient to compel hypertrophic growth in culture (shown by an antisense knockdown of 7SK snRNA), obligatory for hypertrophy in culture (shown by dnCdk9), and sufficient for hypertrophy in mouse myocardium (shown by forced expression of its activator cyclin T1). Findings that implicate Cdk9 as a potential therapeutic target in cardiac disease also include the abnormally increased activity of this kinase in human heart failure and the catastrophic interaction between cyclin T1 and other hypertrophic cues in transgenic mouse models. P-TEFb — the cyclin T/Cdk9 heterodimer — lies at an intriguing interface between the forefront of fundamental studies into general transcription factors, on the one hand, and disease-driven studies of pathobiological growth, on the other. Tantalizing insights are emerging but many questions surrounding its structure, activity, and function remain to be answered.
Consider, first, its endogenous inhibitor, the 7SK snRNA. Recent findings implicate a novel protein, ménage a quatre (MAQ1), as recruited to the N-terminal homology region of cyclin T via this noncoding transcript, a relationship that resembles (and competes with) the binding of cyclin T1 by the HIV Tat protein and HIV TAR RNA. Thus, it has been proposed that lentivirus has subverted the regulation of P-TEFb by the MAQ1–7SK protein-RNA complex (Michels et al., 2003). However, the participation of MAQ1 in growth signaling in vivo and its specific function even in cell culture are still mysteries. Futhermore, even though the 7SK-MAQ1 complex was found to inhibit P-TEFb function, it is unknown how signals release this complex from P-TEFb and which component of the complex is their primary target.
Several cyclins besides T1 partner with Cdk9, including cyclins T2a and T2b, the alternatively spliced products of a separate gene. Like cyclin T1, the T2 cyclins are found in the cardiac P-TEFb/7SK RNA complex but, unlike cyclin T1, both are upregulated highly in the adult heart, compared to myocardium of the embryo or neonate (Sano et al., 2002). Aside from Tat binding just to cyclin T1, functional distinctions among these isoforms are largely conjectures.
Taking a step back from the partnering cyclins, Cdk9 has been shown to associate with other proteins in the cell, such as the molecular chaperones heat shock protein 70 and chaperone complex Hsp90/cyclin-dependent complex 37 (O’Keeffe et al., 2000). Whereas activation of Cdk9 in acute hypertrophic models was traced, thus far, only to the dissociation of 7SK RNA, the chronic hypertrophic models were found to cause not only decreased binding of the inhibitor but also increased expression of the chaperone proteins and increased assembly of cyclin T/Cdk9 (M. Sano and M.D. Schneider, unpublished data). Is this pathway needed for normal activation of Cdk9? Conversely, are Cdk9 activation and its resulting impact on transcription essential targets of these chaperones?
To date, we have learned more of Cdk9 than of Cdk7 in the coupling of hypertrophic signals to RNAPII phosphorylation. However, it should not be overlooked that all chronic hypertrophic signals (long-term load, Gq, and calcineurin) also activated this second of the CTD kinases. Experimentallly, the barriers to studying Cdk7 activation in cardiac hypertrophy are formidable, as no in vitro model has been identified in which its activation is perturbed and activation of Cdk7 in vivo has, thus far, always been concomitant with activation of Cdk9 (Sano et al., 2002). What, then, is the role, if any, of Cdk7? If consequential, how is its activity regulated by the growth signals that also impinge on Cdk9? Conversely, for RNAPII to be competent for recruitment into the preinitiation complex, it must be in the dephosphorylated form. Although critical to the cycling of RNAPII, nothing is presently known about the CTD phosphatases in cardiac hypertrophy.
Although the results discussed here are interpreted in the broadest possible biochemical context — namely, the function of CTD phosphorylation in transcription elongation generically — it can be envisioned that some genes might be regulated more than others by this mode of transcriptional control. As yet, nothing is known of specific genes’ susceptibility to regulation by hypertrophic signals at this step. Such an understanding of CTD kinases in cardiac hypertrophy is likely to require both flavors of scientific "chips" — genome-wide microarray comparisons of gene expression and, more mechanistically, chromatin immunoprecipitation. The latter would be indispensable to pinpoint promoter escape and identify the protein-protein associations that occur with the elongating form of RNAPII provoked by cardiac growth signals.
|
REFERENCES |
|---|
|
TOPABSTRACTI. OverviewII. The RNAPII Phosphorylation...III. Cdk7 and Cdk9 Are Targets of the...IV. Trophic Signals Dissociate an...V. Cdk9 as a Therapeutic TargetVI. Unanswered Questions and Future...REFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Cheng and D. H. Price Properties of RNA Polymerase II Elongation Complexes Before and After the P-TEFb-mediated Transition into Productive Elongation J. Biol. Chem., July 27, 2007; 282(30): 21901 - 21912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Li, J. J. Cooper, G. H. Altwerger, M. D. Feldkamp, M. A. Shea, and D. H. Price HEXIM1 is a promiscuous double-stranded RNA-binding protein and interacts with RNAs in addition to 7SK in cultured cells Nucleic Acids Res., April 3, 2007; 35(8): 2503 - 2512. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Blazek, M. Barboric, J. Kohoutek, I. Oven, and B. M. Peterlin Oligomerization of HEXIM1 via 7SK snRNA and coiled-coil region directs the inhibition of P-TEFb Nucleic Acids Res., December 23, 2005; 33(22): 7000 - 7010. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Li, J. P. Price, S. A. Byers, D. Cheng, J. Peng, and D. H. Price Analysis of the Large Inactive P-TEFb Complex Indicates That It Contains One 7SK Molecule, a Dimer of HEXIM1 or HEXIM2, and Two P-TEFb Molecules Containing Cdk9 Phosphorylated at Threonine 186 J. Biol. Chem., August 5, 2005; 280(31): 28819 - 28826. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Byers, J. P. Price, J. J. Cooper, Q. Li, and D. H. Price HEXIM2, a HEXIM1-related Protein, Regulates Positive Transcription Elongation Factor b through Association with 7SK J. Biol. Chem., April 22, 2005; 280(16): 16360 - 16367. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
