Elsevier

Cellular Signalling

Volume 20, Issue 5, May 2008, Pages 969-977
Cellular Signalling

PKCδ regulates the stimulation of vascular endothelial factor mRNA translation by angiotensin II through hnRNP K

https://doi.org/10.1016/j.cellsig.2008.01.016Get rights and content

Abstract

Angiotensin II (Ang II)-induced renal injury is partly mediated by growth factors such as VEGF. We have previously shown that Ang II rapidly increases VEGF protein synthesis in proximal tubular epithelial (MCT) cells by augmenting mRNA translation, which is partly dependent on activation and binding of hnRNP K to 3′ untranslated region (UTR) of VEGF mRNA. Regulation of hnRNP K activation by PKCδ was studied in MCT cells. Transfection with a PKCδ siRNA inhibited hnRNP K Ser302 phosphorylation and activation, and reduced Ang II stimulation of VEGF synthesis. Inhibition of PKCδ with röttlerin also prevented binding of hnRNP K to VEGF mRNA and reduced the efficiency of VEGF mRNA translation. In db/db mice at 2 weeks of type 2 diabetes, VEGF expression was increased, which was due not to increase in transcription but to augmented translation of VEGF mRNA. Augmented VEGF expression was associated with increased binding of hnRNP K to VEGF mRNA. c-src and PKCδ activities and hnRNP K phosphorylation on Ser302 in renal cortex of db/db mice were increased compared to control mice. We conclude: Ang II-induced VEGF mRNA translation is associated with activation of hnRNP K in MCT cells. In the signaling pathway leading to hnRNP K activation induced by Ang II, PKCδ is downstream of c-src. PKCδ-mediated phosphorylation of hnRNP K is required for Ang II stimulation of VEGF mRNA translation. In mice with type 2 diabetes, src and PKCδ activation and hnRNP K phosphorylation correlate with increased VEGF mRNA translation and kidney hypertrophy. 3′ UTR events are important in regulation of VEGF expression in models of renal injury.

Introduction

Angiotensin II is a mediator of renal injury in diabetic nephropathy and other chronic kidney diseases [1]. Treatment of proximal tubular epithelial cells (MCT cells) with 1 nM of angiotensin II (Ang II) stimulates synthesis of vascular endothelial growth factor (VEGF) through increased translation of its mRNA [2]. This was shown to be due to stimulation of cap-dependent translation initiation, which takes place at the 5′ untranslated region (UTR) of the mRNA [2]. Additional studies showed that VEGF mRNA translation was positively regulated by binding of heterogeneous ribonucleoprotein K (hnRNP K) to the 3′ UTR of the VEGF mRNA [3]. Activation of c-src by Ang II is critical to the binding of hnRNP K to VEGF mRNA, through phosphorylation of the latter on Ser302, which is a target of PKCδ [3].

There are six known tyrosine residues in hnRNP K that undergo phosphorylation by members of the c-src family [4]. Phosphorylation of hnRNP K on tyrosine residues by c-src is believed to generate several c-src homology domain 2 (SH2)-docking sites, which allow the protein to interact with various SH2-containing proteins, such as Lck and PKCδ. Although the tyrosine residues affected in hnRNP K lie outside the KH domains responsible for RNA binding, tyrosine phosphorylation of hnRNP K is generally considered to inhibit its binding to specific RNAs. For instance, Ostareck-Lederer et al. have shown that hnRNP K and the c-src kinase specifically interact with each other, leading to c-src-mediated tyrosine phosphorylation of hnRNP K in vivo and in vitro [4]. c-src-mediated phosphorylation reversibly inhibits the binding of hnRNP K to the differentiation control element (DICE) of the LOX mRNA 3′ untranslated region in vitro and specifically de-represses the translation of DICE-bearing mRNAs in vivo [4]. Similarly, Ostrowski et al. have shown that hnRNP K protein is constitutively bound to many mRNAs in vivo, and that many hnRNP K-mRNA complexes are disrupted by tyrosine phosphorylation [5]. Interestingly, states of enhanced cell proliferation were associated with increased levels of hnRNP K tyrosine phosphorylation [6].

There are four serine residues in hnRNP K that undergo phosphorylation by extracellular signal-regulated kinase (ERK), c-Jun N terminal kinase (JNK), and members of the protein kinase C (PKC) family. It has been shown that serine phosphorylation of hnRNP K is RNA- and DNA-dependent, i.e., RNA- or DNA-bound hnRNP K is a preferred substrate for protein kinases [7]. Serine phosphorylation of hnRNP K is believed to alter its intracellular distribution, leading to cytoplasmic accumulation [8], [9]. Phosphorylation of hnRNP K by ERK also leads to inhibition of mRNA translation [8].

hnRNP K has been shown to act as a docking platform that allows Lck, a member of the c-src family, to interact with a member of another kinase cascade, PKCδ to control the activity of a translation eukaryotic elongation factor, eEF1A (EF-1α) [4], [5], [7], [10]. While bound to RNA, hnRNP K is prevented from direct interaction with PKCδ [10]. The binding of Lck to hnRNP K enhances its activity resulting in phosphorylation of hnRNP K on additional tyrosine residues, causing dissociation of hnRNP K protein from RNA [4], [5] and allowing the recruitment of PKCδ [10]. Following binding, the DAG-primed PKCδ is further induced through tyrosine phosphorylation by the activated Lck docked next to it [10]. Activated PKCδ not only targets Ser302 on hnRNP K [10] but also phosphorylates effectors either bound to K protein or present in its microenvironment through imposed proximity. EF-1α binds hnRNP K [11] and is a substrate of PKCδ [12]. Therefore, PKCδ-mediated phosphorylation of EF-1α could occur in the context of hnRNP K, and contribute to the activation of the translational machine.

Although the importance of hnRNP K phosphorylation by PKCδ has been established [3], [10], the functional consequences of phosphorylation of hnRNP K by PKCδ are not known. Also unknown is whether similar association between hnRNP K and PKCdelta occurs in vivo in the kidney in a disease state. In this study, we sought to investigate the role of PKCδ-mediated phosphorylation of hnRNP K in the regulation of VEGF mRNA translation by Ang II in MCT cells and assess the association between the two proteins in vivo in the kidney tissue of mice with type 2 diabetes.

Section snippets

Cell culture

SV40-immortalized murine proximal tubular epithelial cells (MCT) were provided by Dr. Eric Neilson, Vanderbilt University, Nashville, TN. MCT cells in culture express in vivo characteristics of proximal tubular epithelial cells. [13]. The cells were grown in Dulbecco's minimal essential medium (DMEM) containing 5 mM glucose and 10% FBS [2], [3]. Confluent monolayers of cells were serum-deprived in DMEM for 18 h before treatment.

Animal experiments

C57BLKsJ lepr−/− db/db mice and their lean littermate controls

Results

Ang II stimulates PKCδ activity in MCT cells with a time-course corresponding to that of increased VEGF synthesis [3]. Phosphorylation of hnRNP K on Ser302, a known target of PKCδ [10], positively correlated with its binding to VEGF mRNA, and knockdown of hnRNP K expression by RNA interference significantly reduced Ang II stimulation of VEGF synthesis [3]. However, the requirement for PKCδ for either hnRNP K phosphorylation on Ser302 or for Ang II-induction of VEGF synthesis has not been

Discussion

Our data demonstrate the following: (1) Inhibition of PKCδ expression reduces Ang II stimulation of VEGF synthesis, similar to inhibition of hnRNP K expression. (2) hnRNP K recruits both c-src and PKCδ activated by Ang II that allows phosphorylation and activation of PKCδ by c-src. (3) PKCδ, in turn, phosphorylates and activates hnRNP K, allowing its binding to VEGF mRNA (4) Following Ang II stimulation, both hnRNP K binding to VEGF mRNA and its translation are PKCδ dependent. (5) In kidney

References (40)

  • J. Ostrowski et al.

    J. Biol. Chem.

    (2000)
  • D.S. Schullery et al.

    J. Biol. Chem.

    (1999)
  • K. Bomsztyk et al.

    FEBS Lett.

    (1997)
  • W. Lu et al.

    Biochem. Biophys. Res. Commun.

    (2007)
  • M. Gschwendt et al.

    Biochem. Biophys. Res. Commun.

    (1994)
  • D. Senthil et al.

    Kidney Int.

    (2003)
  • Z. He et al.

    J. Biol. Chem.

    (2005)
  • K. Utsugisawa et al.

    Brain Res. Mol. Brain Res.

    (2005)
  • A. Policha et al.

    J. Biol. Chem.

    (2006)
  • J.R. Giorgione et al.

    J. Biol. Chem.

    (2006)
  • H. Ha et al.

    Am. J. Kidney Dis.

    (2001)
  • H. Rincon-Choles et al.

    Kidney Inter., Suppl.

    (2002)
  • D. Feliers et al.

    Am. J. Physiol., Renal Physiol.

    (2005)
  • D. Feliers et al.

    Am. J. Physiol., Renal Physiol.

    (2007)
  • A. Ostareck-Lederer et al.

    Mol. Cell. Biol.

    (2002)
  • J. Ostrowski et al.

    Br. J. Cancer

    (2003)
  • I. Van Seuningen et al.

    Biochemistry

    (1995)
  • H. Habelhah et al.

    Nat. Cell Biol.

    (2001)
  • M.D. Koffa et al.

    J. Virol.

    (2003)
  • K. Kielbassa et al.

    J. Biol. Chem.

    (1995)
  • Cited by (28)

    • A novel role for the nuclear localization signal in regulating hnRNP K protein stability in vivo

      2016, Biochemical and Biophysical Research Communications
      Citation Excerpt :

      These modifications include methylation [17] and phosphorylation by ERK [18] of key residues within an RGG motif and nuclear shuttling domain (KNS), respectively. Once in the cytoplasm, hnRNP K initially represses translation [16] until it undergoes additional post-translational modifications, such as phosphorylation by Src [19], JNK [6], ERK [7], and PKC-delta kinases [20]. As translation is activated, a small portion of cytoplasmic hnRNP K remains associated with polyribosome-bound, actively translated mRNAs [6,21].

    • Phosphorylation of heterogeneous nuclear ribonucleoprotein K at an extracellular signal-regulated kinase phosphorylation site promotes neurofilament-medium protein expression and axon outgrowth in Xenopus

      2015, Neuroscience Letters
      Citation Excerpt :

      Once in the cytoplasm, these mRNAs remain translationally repressed within hnRNPK-mRNP complexes until additional signals promote translation. Our study adds ERK to the list of kinases promoting this translation (i.e., JNK [10], c-Src [19] and PKC-δ [21]) and further implicates hnRNPK as a likely target of ERK signaling promoting axon outgrowth. Since ERK also regulates NF transport dynamics [27], neurons may use ERK signaling to help coordinate NF transport with synthesis.

    • Transcriptional regulation of heterogeneous nuclear ribonucleoprotein K gene expression

      2015, Biochimie
      Citation Excerpt :

      Bomsztyk K et al. found hnRNP K was bound to elongation factor 1α (EF-1α) [12], and a series of studies showed a role of hnRNP K in protein translation. The tyrosine and Ser302 of hnRNP K were phosphorylated by angiotensin II through Ang II-AT1R-src-PKCδ pathway, which promoted the translation of VEGF mRNA [13,14]. Together with hnRNP E1and DEAD-box RNA helicase 6(DDX6), hnRNP K can also bind to the differentiation control element (DICE) in 3′ untranslated region (3′UTR) of 15-lipoxygenase mRNA to form a silencing complex, which could inhibit the translation of 15-lipoxygenase by control the assembly of 80s ribosomes [15,16].

    • Aurora-A phosphorylates hnRNPK and disrupts its interaction with p53

      2011, FEBS Letters
      Citation Excerpt :

      HnRNPK is composed of three K homology (KH) domains responsible for DNA/RNA binding and one K-interactive (KI) region for protein-protein interactions (Fig. 1). Several post-translational modifications of hnRNPK have been shown to regulate its DNA-binding, translational regulation, localization, and protein-protein interaction [18–24]. In this study, we demonstrated that Aurora-A directly interacts with and phosphorylates hnRNPK on Ser-379 in vitro and in vivo.

    • Mechanism of VEGF expression by high glucose in proximal tubule epithelial cells

      2010, Molecular and Cellular Endocrinology
      Citation Excerpt :

      Monolayers of MCTs were grown to 60–70% confluence and serum-deprived overnight before treatment, and were barely confluent at the time of treatment. Immunoblotting experiments were performed as previously described (Feliers et al., 2007; Sataranatarajan et al., 2008). MCTs were homogenized in lysis buffer (50 mM Tris·HCl, pH 7.4, 150 mM KCl, 1 mM EDTA, 50 mM β-glycerophosphate, 0.1 mM sodium orthovanadate, 1 mM EGTA, 0.5% Nonidet P40, and protease inhibitor mix [Sigma, St. Louis, MO]).

    View all citing articles on Scopus
    View full text