Role of p90^RSK in regulating the Crabtree effect: implications for cancer

Abstract

High glucose inhibits mitochondrial respiration, known as the “Crabtree effect”, in cancer cells and possibly other cell types. The upstream pathways regulating this phenomenon are poorly understood. In diabetes, where glucose levels are elevated, the p90 ribosomal S6 kinase (p90^RSK) has received much attention as a potential upstream mediator of the effects of high glucose. Evidence is also emerging that p90^RSK may play a role in cancer cell signaling, although the role of p90^RSK in regulating cancer cell metabolism is unclear. Herein we provide an overview of the Crabtree effect and its relationship to mitochondrial metabolism. Furthermore, preliminary data are presented suggesting a role for p90^RSK and its upstream components, the ERK family of mitogen-activated protein kinases (MAPKs), in the Crabtree effect.

Introduction

The exposure of cells to high concentrations of glucose results in a number of physiological and pathological responses. In the field of diabetes research, the effects of glucose on mitochondrial reactive oxygen species (ROS) generation have been well-studied, and are believed to underlie much of the secondary pathology of diabetes, particularly vascular complications [1]. In the field of cancer research, the effects of glucose on cancer cell metabolism have also been studied extensively, but the majority of attention has been given to the Warburg effect – the phenomenon in which cancer cells appear to prefer anaerobic glycolysis as a source of ATP even in the presence of adequate O₂ for mitochondrial oxidative phosphorylation (Ox-Phos) to occur [2,3].

In contrast the Crabtree effect, postulated in 1929, has languished in the shadow of the Warburg effect in terms of research attention [4]. The Crabtree effect is simply defined as the observation that high concentrations of glucose result in a decrease in mitochondrial respiration in cancer cells [4]. The role of inhibited mitochondrial function in the Warburg effect is still unclear [5], and as such it is possible that the Crabtree effect may be an underlying cause of the Warburg effect. Recently, it has become apparent that the Crabtree effect is present in many types of dividing cells (e.g. endothelial cells), and not just in cancer cells [6]. Furthermore, it may be related to the phenomenon of glucose repression which is observed in yeast [7]. Although it was discovered eight decades ago, the molecular mechanisms underpinning the Crabtree effect have yet to be elucidated.

Recent evidence has implicated the p90 ribosomal S6 kinase (p90^RSK) in glucose sensing and diabetes [8]. Thus, it is possible that p90^RSK may play a role in the Crabtree effect. The current paper presents evidence favoring a role for p90^RSK in the Crabtree effect, and discusses this in the context of cancer signaling.

ERK5/MEK5/p90^RSK pathway in cancer

Mitogen-activated protein kinases (MAPKs) are essential components many cell signaling events including proliferation, differentiation and cell death. The biological actions of ERK5 (Big MAP Kinase 1) are well described [9,10] in several cell proliferative and developmental pathways, as well as cancer. MEK5 [10,11] has been closely linked to ERK5, and nearly all cellular processes in which MEK5 has been implicated, are attributed to regulation by ERK5.

Within the field of cancer research, the MEK5-ERK5 pathway has been implicated in cell growth control [12], and a role for ERK5 has been postulated in EGF-induced proliferation and cell-cycle progression [13]. Furthermore, studies in ERK5 knock-out mice show a crucial role for ERK5 in angiogenesis and vascular cell growth [14], especially as a modulator of other antigenic signals.

In patients suffering from multiple myeloma, ERK5 is expressed in B-cells [15] and it is constitutively activated in breast cancer cells over-expressing the Erb B2 receptor [16]. ERK5 activation is required for the proliferation of myeloma and breast cancer cell lines in response to IL-6 and neuregulins respectively. Cyclin D1, whose deregulation is implicated in a wide variety of cancers, was identified as a target of the MEK5/ERK5 pathway [17]. The authors showed that the kinase activity of ERK5 is required for induction of cyclin D1 in response to serum in some breast cancer cell lines [17]. In prostate and breast cancers, MEK5 over-expression is associated with poor survival and resistance to chemotherapy respectively [18,19]. Furthermore, prostate cancer cell growth can be suppressed by inhibition of MEK5 signaling, through the enhanced degradation of the protein [20]. ERK5 and p90^RSK have some overlapping functions in cancer: p90^RSK, like ERK5, is thought to stimulate cell cycle progression [21]. ERK5 also responds to cAMP [22] and plays a role in signaling to the pro-apoptotic protein Bad [23], which is a direct substrate of p90^RSK [24].

Overall, there are several lines of evidence linking p90^RSK to cancer and to glucose sensing, but surprisingly the role of this serine/threonine kinase in mediating the Warburg or Crabtree effects has not been investigated.

Experimental

The metabolic function of the H9c2 cardiomyocyte cell line was measured using an XF24 extracellular flux analyzer (Seahorse Bioscience, Billerica MA) [25,26]. Cells were incubated in a glucose free Krebs Ringer solution (pH 7.3 w. HEPES) for 30 min., with or without 20 mM glucose, with the addition of kinase inhibitors and other pharmacologic agents, as follows: (i) BIX02188 (MEK5 inhibitor, 20 μM); (ii) UO126 (MEK1/2 inhibitor, 10 μM); (iii) FMK (fluoromethylketone pyrrolopyrimidine, p90^RSK inhibitor [21], 10 μM); (iv) Splitomicin (SIRT1 inhibitor, 10 μM).

Figure 1 shows that H9c2 cells exhibited a robust Crabtree effect, with 30 min. exposure to 20 mM glucose resulting in a 20% drop in cellular oxygen consumption rate. This data is in broad agreement with those of Sweet et al. [6], who reported a robust Crabtree effect in endothelial cells. Thus, the Crabtree effect is not unique to cancer cells, and may apply to many types of dividing cells. Notably, this effect was blocked by pre-incubation with the p90RSK inhibitor FMK (Fig. 1A) or the MEK5 inhibitor BIX (Fig. 1B). Other pharmacophores tested included the MEK1/2 (ERK1/2) inhibitor UO126, or the SIRT1 inhibitor splitomicin, but neither had a significant effect on the inhibition of respiration by glucose (data not shown).

Figure 1. Crabtree Effect in H9c2 Cells, Effect of Inhibitors.

Oxygen consumption of H9c2 cells was measured in glucose free or replete (20 mM) media, in the presence or absence of the p90^RSK inhibitor FMK (panel A), and the MEK5 inhibitor BIX (panel B). Data are means ± SEM, N >3. See text for further details.

Discussion

The role of MAP kinases in the response to glucose in general has been well-studied, and it was shown that some mitochondrial responses to high glucose could be blocked by high concentrations of the non-specific MEK1/2 (ERK1/2) inhibitor PD98059 [27], although this effect may have been due to off-target effects of PD98059 [28].

More recently, attention within the MAPK field has focused on another member of the ERK family, ERK5 (Big MAP Kinase), which is an upstream mediator of p90^RSK activity [29,30]. This has been facilitated by the commercial availability of BIX02188, an inhibitor of MEK5, which is the kinase immediately upstream of ERK5. Our data in Fig. 1B suggest that indeed MEK5, through ERK5, is an upstream kinase involved in the activation of p90^RSK in response to high glucose. This signaling pathway is shown in Figure 2, suggesting key roles for both MEK5 and p90^RSK in the Crabtree effect.

Figure 2.

Proposed Signaling Pathway Linking High Glucose to Respiratory Inhibition.

To the best of our knowledge, this is the first identification of any signaling pathway involved in mediating the Crabtree effect. Notably, p90^RSK has been linked to signaling by the tumor suppressor protein p53 [31]. In addition, p53 is emerging as an important regulator of mitochondrial function [32–34]. Thus, p90^RSK may play additional roles in cancer signaling beyond the Crabtree effect, as reviewed above.

The relationship between the Crabtree effect and mitochondrial morphology is also worthy of further consideration. Specifically, a link has been established between mitochondrial fragmentation and ROS generation in response to high glucose [27,35]. Thus it is interesting to speculate that mitochondrial fragmentation may be an intermediate step between p90^RSK and the inhibition of respiration. Clearly, the elucidation of this complex signaling pathway which drives the response of cancer and other dividing cell types to high glucose, may lead to the development of novel therapeutic approaches, based on the inhibition of p90^RSK and its upstream signals.