TRPM7 ion channels are required for sustained phosphoinositide 3-kinase signaling in lymphocytes

Summary

Lymphocytes lacking the TRPM7 dual function ion channel/protein kinase exhibit a unique phenotype: they are unable to proliferate in regular media, but proliferate normally in media supplemented with 10–15 mM extracellular Mg²⁺. Here, we have analyzed the molecular mechanisms underlying this phenotype. We find that upon transition from proliferation-supporting Mg²⁺-supplemented media to regular media, TRPM7-deficient cells rapidly downregulate their rate of growth, resulting in a secondary arrest in proliferation. The downregulated growth rate of transitioning cells is associated with a deactivation of signaling downstream from phosphoinositide 3-kinase, and expression of constitutively active p110 phosphoinositide 3-kinase is sufficient to support growth and proliferation of TRPM7-deficient cells in regular media. Together, these observations indicate that TRPM7 channels are required for sustained phosphoinositide 3-kinase-dependent growth signaling, and therefore that TRPM7 is positioned alongside phosphoinositide 3-kinases as a central regulator of lymphocyte growth and proliferation.

Introduction

TRPM7 (transient receptor potential cation channel, subfamily M, member 7) is a novel dual function protein that possesses both ion channel and protein kinase domains. TRPM7 channel function has been shown to be responsive to multiple receptor-mediated inputs (reviewed in (Schmitz et al., 2007)), and studies from our laboratory have shown that TRPM7-deficient cells fail to proliferate in normal tissue culture media (Schmitz et al., 2003), suggesting a role for TRPM7 channel function in regulation of cell proliferation. Remarkably, TRPM7-deficient cells are able to proliferate normally when placed in media with 10–15 mM supplemental extracellular Mg²⁺ (physiological [Mg²⁺]_ex= 0.4 mM). In conjunction with the observation that humans deficient in TRPM6, a close TRPM7 homologue predominantly expressed in the kidney, exhibit a specific defect in renal Mg²⁺ uptake, complementation of the TRPM7-deficient proliferative arrest phenotype by supplemental extracellular Mg²⁺ or heterologous Mg²⁺ transporter expression has led to the proposal that TRPM7 is involved in regulating Mg²⁺ uptake required for cell proliferation (Sahni et al., 2007; Schmitz et al., 2004; Schmitz et al., 2003). However, the molecular mechanisms leading to the proliferative arrest of TRPM7-deficient cells and how these events are related to cellular Mg²⁺ uptake have remained undefined.

To gain further insight into TRPM7's role in supporting proliferation, we analyzed TRPM7-deficient lymphocytes during their transition to proliferative arrest in regular tissue culture media. Cell cycle analyses showed that TRPM7-deficient cells cultured in regular media accumulate in the G₀/G₁ and early G₂ phases of the cell cycle, suggesting that their proliferative arrest is due to a primary failure of cell growth (increase in size and mass), as opposed to a failure of DNA synthesis or mitosis. Biochemical and genetic analyses demonstrated that the growth failure of TRPM7-deficient cells cultured in regular media is due to a lack of sustained activation of phosphoinositide 3-kinase dependent signaling, including growth regulatory signaling along the mTORC1 pathway. Overall, our results suggest that TRPM7 regulates lymphocyte growth through a requirement of its channel function for sustained activation of phosphoinositide 3-kinase growth regulatory pathways.

Results

TRPM7-deficient cells fail to grow

Inhibition of the proliferation of tumor cells in culture may be induced by compromise of biosynthetic metabolism, inhibition of DNA synthesis, DNA damage, or inhibition of mitosis. These mechanisms are distinguishable through analysis of DNA content and cell size (reviewed in (Cooper, 2004)). To determine which of these mechanisms contribute to the proliferative arrest of TRPM7-deficient cells cultured in regular media, we analyzed the DNA content (via DAPI staining) and cell size (via forward scatter) of TRPM7-deficient cells upon transition from proliferation in growth supporting Mg²⁺-supplemented media to regular media (Figure 1A). After 24 hours in regular media, TRPM7-deficient cells displayed a striking accumulation of cells in the G₀/G₁ phase of the cell cycle (left panel), as well as a reduction in average cell size (middle panel). Furthermore, comparison of the size of WT cells and TRPM7-deficient cells with 2n or 4n DNA content indicated that TRPM7-deficient cells arrested in G₀/G₁ (2n content) are smaller than those arrested in G₂ (4n content), and that G₂-arrested TRPM7-deficient cells are smaller than a significant fraction of WT cells, consistent with an arrest in cell growth occurring at a cell's relative cell cycle position at the time of transition (right panel). Taken together, these results indicate that TRPM7-deficient cells undergo a failure of growth – i.e. increase in size and mass - as opposed to primary failure or compromise of DNA replication, DNA damage, or mitosis – in conjunction with their proliferative arrest in standard cell culture conditions.

Figure 1. Proliferative arrest of TRPM7-deficient cells in regular media is due to primary failure of an endogenous growth regulatory mechanism.

A – Proliferative arrest of TRPM7-deficient cells is associated with reduced average cell size and accumulation of cells in the G₀/G₁ cell cycle phase. Left panel: DNA content analysis of cell cycle distribution of WT and TRPM7-deficient cells after 24 hours of culture in regular tissue culture media (−) or regular media supplemented with 15 mM Mg²⁺ (+). Middle panel: Forward scatter analysis of cell size distribution of DT40 WT and TRPM7-deficient cells after the indicated periods of culture in regular tissue culture media (−) or regular media supplemented with 15 mM Mg²⁺ (+). Right panel: Forward scatter analysis of cell size distribution of TRPM7-deficient cells with 2n or 4n DNA content after the indicated periods of culture in regular tissue culture media (−).

B – WT DT40 cells cultured under conditions in which Mg²⁺ uptake is specifically compromised exhibit reduced cell size and G₀/G₁ cell cycle accumulation. Left panel: DNA content analysis of cell cycle distribution of WT DT40 cells after 24 hours of culture in depolarizing tissue culture media (DP) with the indicated amounts of added Mg²⁺ (as MgCl₂). Right panel: Forward scatter analysis of cell size distribution of WT DT40 cells after 24 hours of culture in depolarizing tissue culture media (DP) with the indicated amounts of added Mg²⁺ (as MgCl₂).

All results are representative of three separate experiments.

As our TRPM7-deficient cells were generated via gene targeting in the presence of media with supplemental Mg²⁺, it is possible that their growth failure reflects a non-physiological response to the decreased extracellular Mg²⁺ concentrations present in standard tissue culture media. To evaluate WT DT40 cells’ response to limitation of Mg²⁺ uptake, we placed WT DT40 cells into culture conditions in which both electrical (membrane potential) and chemical (concentration gradient) forces for Mg²⁺ entry were compromised (Figure 1B). In these experiments, WT DT40 cells were placed in Mg²⁺-free tissue culture media in which 95 mM NaCl was replaced with 95 mM KCl (thereby substantially reducing their membrane potential – this media is designated as depolarizing media or DP), and with varying amounts of added Mg²⁺ (as MgCl₂). After 24 hours in these conditions, cell cycle and size were analyzed as in Figure 1B. As can be seen, WT DT40 cells are able to proliferate normally and exhibit a normal cell size when grown in depolarizing media with 0.4 mM MgCl₂ added (the same amount present in typical RPMI-1640 tissue culture media): thus the reduced membrane potential achieved by the depolarizing media does not compromise nutrient uptake to a point that growth is affected. However, as extracellular Mg²⁺ is progressively decreased, the WT DT40 cells undergo a similar type of growth arrest to that observed in the TRPM7-deficient cells grown in regular media, with accumulation of cells in G₀/G₁ and G₂ (left panel) along with a progressive reduction in average cell size (right panel). Furthermore, cells with 2n had smaller sizes than those with 4n DNA content, as observed above for TRPM7-deficient cells cultured in regular media (data not shown). Overall, these results indicate that the growth failure and cell cycle changes observed in TRPM7-deficient cells placed in regular media are the result of a cellular regulatory response to limitation of Mg²⁺ uptake.

Translational down-regulation in TRPM7-deficient cells

Eukaryotic cells have the capacity to adjust their rate of growth in response to nutrient limitations (amino acids, glucose, etc.) and growth factors by regulating macromolecular synthesis through a well defined biochemical mechanism involving the signaling complex mTORC1 (mammalian target of rapamycin complex 1, reviewed in (Wullschleger et al., 2006)). As the failure of cell growth observed in TRPM7-deficient cells resembled that induced by the mTORC1 inhibitor rapamycin, we assessed the activation status of mTORC1 signaling in transitioning TRPM7-deficient cells by interrogating a key effector in mTORC1-dependent regulation of cell growth, ribosomal S6 kinase (S6K). S6K, which is expressed in two forms in mammalian cells, S6K1 and S6K2, regulates ribosomal translation through phosphorylation of the S6 ribosomal subunit, and is thought to be a direct downstream target of mTORC1. The activation status of S6K1 in TRPM7-deficient cells transitioned between growth supporting Mg²⁺-supplemented media and growth arresting regular media was assessed using immunoblotting analyses of the phosphorylation level of Thr389 of S6K1, an established indicator of its activation status (Shah and Hunter, 2004) with or without 100nM rapamycin. As can be seen in Figure 2A (top left panel), S6K1 Thr389 phosphorylation is detectably down-regulated by 3 hours post-transfer of TRPM7-deficient cells from growth supporting to regular media, and substantially down-regulated by 5 hours post-transfer. Similarly, transfer of TRPM7-KO (knock out) cells cultured for 5 hours in regular media to growth supporting conditions resulted in a marked upregulation in S6K1 phosphorylation which was significantly detectable by 2 to 4 hours (top middle and right panels). As expected, rapamycin treatment in both cases inhibited p-S6K1. That this response reflects a normally operative endogenous regulatory pathway was evaluated by placing WT DT40 cells in Mg²⁺ free depolarizing media and evaluating S6K1 phosphorylation before and after re-addition of Mg²⁺ - WT DT40 cells exhibited a similar reduction of S6K1 activation over five hours exposure to Mg²⁺ free depolarizing media (Figure 2A, bottom panel, compare lane 1 with lane 5), which was rapidly reversed upon provision of physiological amounts of Mg²⁺ (Figure 2A, bottom panel, compare (+) to (−) media in lanes 1–4).

Figure 2. Growth arrest of TRPM7-deficient cells correlates with failure of growth regulatory signaling above the level of mTOR protein kinase.

A – Growth arrest of TRPM7-deficient and WT DT40 cells correlates with deactivation of the growth regulator effector ribosomal S6 kinase (S6K1). Upper left panel: TRPM7-KO cells were washed twice with PBS and transferred to fresh media with supplemental Mg²⁺ for 2 hours, following which the cells were washed and transferred to regular media without (−) or with (+) supplemental Mg²⁺ and analyzed for p-S6K1 (Thr389) activation. Cells pre-treated with 100nM rapamycin (Rap) displayed a significant reduction in p-S6K1. S6K1 showed a detectable decrease in phosphorylation by 3 hours in regular media alone, and was substantially down-regulated by 5 hours. Upper middle panel: TRPM7-KO cells were transferred to fresh media without any supplemental Mg²⁺ (−) for 5 hours before transitioning the cells to regular media without (−) or with (+) 15mM supplemental Mg²⁺. P-S6K1 (Thr389) showed an increase in activity within 2 hours of transition to (+) media conditions. Upper right panel: Analysis of the fold change in p-S6K1 (Thr389) from three independent experiments. A 4-fold increase was seen by 2hrs, increasing slightly further to 5-fold by 4hrs respectively, in cells growing under +media conditions vs. – media conditions. P values for the fold change were calculated using Student’s t-test and changes at 2 hrs and 4 hrs were found to be highly statistically significant (p=0.002 for 2hrs and p=0.016 for 4hrs). Lower middle panel: WT DT40 cells were placed in Mg²⁺-free depolarizing media for 5 hours, and then 0.4 mM Mg²⁺ was added back at the indicated times. Alternatively, WT DT40 cells were maintained in regular media for the duration of the experiment as indicated.

B – TRPM7-deficient cells exhibit deactivation of the growth regulatory effector 4E-BP1 but not the mitogen activated kinase p42 ERK. Left panel: Activity of phospho-4E-BP1 (Ser65) was slightly increased by 2–4 hours after Mg²⁺ add-back and was inhibited in presence of 100nM rapamycin. Right panel: p-Erk (Thr185/Tyr187) remained unchanged under experimental conditions mentioned above, in absence (−) or presence (+) of 15mM supplemental Mg²⁺ in the growth media. Note - the top band observed in the total Erk blot is thought to be non-specific as DT40 cells only express Erk2.

C – TRPM7-deficient and WT DT40 cells exhibit G₀/G₁ arrest in response to the mTOR inhibitor rapamycin which is not reversible via supplemental Mg²⁺. Left panel: Rapamycin (100nM) induces a G₁ arrest in DT40 WT cells both in absence (−) and presence (+) of supplemental Mg²⁺ at 48 hours. Middle panel: Rapamycin induces a G₀/G₁ arrest in the TRPM7-KO cells by 48 hours, which is not reversible by supplemental Mg²⁺. Right panel: Cell size analysis of DT40 WT and TRPM7-KO cells in presence of 15mM Mg²⁺ and rapamycin. Both DT40 WT cells as well as the TRPM7-KO cells show a reduction in cell size in presence of rapamycin and 15mM supplemental Mg²⁺ which is comparable to the cell size of TRPM7-KO cells growing in normal medium with 0.4mM Mg²⁺.

All data are representative of three independent experiments.

We also evaluated two other proteins implicated in regulation of cell growth and proliferation: 4E-BP1 (eukaryotic translation initiation factor 4E binding protein 1), which like S6K1, is thought to be a major direct target of the mTORC1 complex (Bhaskar and Hay, 2007), and p42 ERK, a mitogen activated protein kinase (MAPK) and an important effector of mitogenic signaling activated by growth factor receptors (Dhillon et al., 2007). In TRPM7-deficient cells cultured for 5 hours in regular media, 4E-BP1 exhibited slightly reduced Ser65 phosphorylation, as well as reduced abundance of its active high molecular weight form (Supplemental Data, Figure S1, right panel, last 3 lanes). 4E-BP1 also exhibited re-phosphorylation of Ser65 and an increased abundance of its active high molecular weight form after transfer from regular media to growth supporting conditions (Figure 2B, left panel), supporting the hypothesis that mTORC1 growth regulatory signaling is diminished in TRPM7-deficient cells cultured in regular media. After treatment with rapamycin (100nM), the phosphorylation level of 4E-BP1 was drastically reduced as expected. We were unable to detect changes in the phosphorylation of Erk (Thr185/Tyr187) within a 4 hour time frame after transition of cells to growth supporting conditions (Figure. 2B, right panel), consistent with TRPM7-deficient cells continuing to receive a full complement of mitogenic signaling from the tissue culture media.

mTORC1 activation fails in TRPM7-deficient cells

Taken together, the above results indicate that TRPM7-deficient cells transitioned to regular media undergo a downregulation of cell growth, which is closely correlated, with interruption of growth signaling to downstream effectors S6K1 and 4E-BP1. As mentioned above, the mammalian target of rapamycin (mTOR) protein kinase component of the mTORC1 complex is thought to directly phosphorylate and activate S6K1 and 4E-BP1, and is highly specifically inhibited by the macrolide antibiotic rapamycin. To establish whether TRPM7-dependent regulation of S6K1 and 4E-BP1 occurs at or above the level of mTORC1, we treated WT and TRPM7-KO cells with 100nM rapamycin, cultured them in regular media or Mg²⁺ supplemented growth supporting media, and analyzed their cell cycle distribution and cell size (Figure 2C). As seen in Figure 2C, left panel, rapamycin induced an accumulation of WT DT40 cells in G₀/G₁ by 48hrs, but the observed accumulation was not reversed upon culturing the WT DT40 cells in growth supporting media with supplemental Mg²⁺. A 48 hour treatment with rapamycin induced a similarly significant accumulation in G₀/G₁ of TRPM7-deficient cells grown in growth supporting media with supplemental Mg²⁺ (Figure 2C, middle panel), and led to a reduction in cell size of both WT DT40 and TRPM7-deficient DT40 cells (Figure 2C, right panel). Culture in growth supporting Mg²⁺ supplemented media was also not able to prevent the rapamycin-induced reduction in cell size in either TRPM7-KO or wild-type cells (Figure. 2C, right panel). Overall, the lack of ability of Mg²⁺ supplementation to reverse the effects of rapamycin in WT or TRPM7-deficient DT40 cells indicates that the downregulation of growth regulatory signaling, which occurs when TRPM7-KO DT40 cells are cultured in regular media, is caused by an interruption of growth regulatory signaling at or above the level of mTORC1.

TRPM7-deficient cells do not sustain Akt activation

The mTORC1 growth regulatory complex is regulated by upstream signals from growth factor receptors and nutrient-sensing pathways relayed through the membrane proximal effectors Akt and phosphoinositide 3-kinase (Bhaskar and Hay, 2007). In order to understand whether signals upstream of mTORC1 are influenced when TRPM7-deficient cells are transitioned to regular media, we assessed the activation of Akt using a similar approach to that described earlier for S6K1: cells were transitioned to regular media for 5 hours, followed by re-transition back to growth supporting Mg²⁺ supplemented media (time of retransition = time 0 for these experiments). The activation state of Akt was then assessed by immunoblotting with phospho-specific antibodies to Ser473 and Thr308, phosphorylation of which have been implicated in Akt activation (Woodgett, 2005). Analogous to p-S6K1, phosphorylation of Akt (Ser473) was substantially reduced in TRPM7-deficient cells cultured for 5 hours in regular media (Figure S2, right panel). Similarly, phosphorylation of two sites implicated in Akt activation - Ser473 and Thr308, were detectably augmented as early as 1hr after transition back to growth supporting media, indicating that mTORC1 deactivation in TRPM7-deficient cells is at least in part due to a failure of growth factor and nutrient signals relayed through Akt (Figure 3A, left and middle panels). Rapamycin treatment did not have an inhibitory effect on Akt phosphorylation (see indicated lanes in both panels), consistent with Akt’s proposed position upstream from mTORC1 in the mTORC1 growth regulatory signaling pathway. A comparative phospho-Akt (S473) analysis between DT40 WT and TRPM7-deficient cells under optimal growth conditions showed similar levels of phosphorylation at 4 hrs (Figure 3A, right panel), demonstrating that the observed changes are not due to idiosyncratic differences in Akt levels or activation between the cell lines.

Figure 3. Growth arrest of TRPM7-deficient cells in regular media correlates with failure of sustained Akt activation.

A – Growth arrest of TRPM7-deficient cells in regular media correlates with failure of Akt activation. Top left panel: Immunoblotting analysis of Ser473 phosphorylation of Akt after transition of TRPM7-deficient cells from regular media (−) to the presence of supplemental Mg²⁺ (+). TRPM7-deficient cells were placed in regular media for 5 hours, and continued in those conditions (−) or 15mM supplemental Mg²⁺ was added (+) for the indicated period of times, followed by cell lysis, SDS-PAGE, and western immunoblotting analysis of Ser473 phosphorylation. Data are representative of two independent experiments. Lower left panel: Western immunoblot analysis of Thr308 phosphorylation of Akt in an identical single experiment. Right panel: A comparative phospho-Akt (S473) analysis between DT40 WT and TRPM7-deficient cells under optimal growth conditions (+) and regular media (−) at 4 hrs. Data from one of three independent experiments are shown.

B – Insulin stimulation of TRPM7-deficient cells induces transient Akt activation in regular media. Left panel: Immunoblotting analysis of Ser473 phosphorylation of Akt in response to transition from regular media (−) to Mg²⁺ supplemented media (+) or in response to insulin stimulation of WT DT40 cells. WT DT40 cells were placed in regular media (−) for 2 hours, and then maintained in that media (−), or 15mM supplemental Mg²⁺ (+) or 100 nM insulin was added for the indicated period of time. Cells were then lysed and analyzed by SDS-PAGE and western immunoblotting for changes in Akt Ser473 phosphorylation. Right panel: Immunoblotting analysis of Ser473 phosphorylation of Akt after insulin stimulation of TRPM7-deficient cells in regular (−) or growth supporting Mg²⁺ supplemented media (+). TRPM7-deficient cells were placed in regular media for 5 hours, and then maintained in that media with 100 nM insulin (−) or media with 100 nM insulin and 15 mM supplemental Mg²⁺ (+) for the indicated period of time. Cells were then lysed and analyzed by SDS-PAGE and western immunoblotting for changes in Akt Ser473 phosphorylation. These results are representative of two independent experiments.

A potential explanation for the above results is that growth factors, such as insulin, are simply unable to activate Akt in the TRPM7-deficient cells grown in regular medium. To evaluate this possibility, we analyzed insulin-dependent Akt activation (as assessed by Ser473 phosphorylation, e.g. see (Pogue et al., 2000)). In WT DT40 cells, starting from an already significant basal level, Akt phosphorylation is not influenced by the presence of supplemental Mg²⁺, but addition of 100 nM insulin generates a readily detectable increase at 1 hour post-addition (Figure 3B, left panel). In contrast, for TRPM7-deficient cells, at 1 hour post-stimulation, Akt phosphorylation is comparably enhanced by either supplemental Mg²⁺ or the addition of 100 nM insulin (Figure 3B, right panel). However, Akt activation decays back to its time 0 baseline by 2 hrs in insulin-stimulated TRPM7-deficient cells cultured in regular media. That insulin affects other aspects of the insulin-receptor signaling pathway equivalently in the two cell types was confirmed by the observation of identical insulin-induced downregulation of surface insulin receptors by western blotting (Figure S2, left panel). Thus, TRPM7-deficient cells grown in regular media are competent for transducing acute Akt activating signals from growth factors such as insulin, yet are unable to maintain sustained Akt phosphorylation.

Constitutive PI3K supports TRPM7-independent growth

The above results suggest a potential causal role for deficient sustained Akt activation in the failure of TRPM7-deficient cells to grow in regular media. To determine whether a heterologous signal providing sustained Akt activation could drive TRPM7-deficient cells to proliferate in regular media, we generated a doxycycline-regulated construct expressing myristoylated Akt (myr-Akt) lacking a PH (pleckstrin homology) domain. However, while we observed complete biochemical complementation at the level of Akt and S6K1 activation, myr-Akt was not able to support proliferation of TRPM7-deficient cells (see Figure S3).

The failure of sustained Akt activation observed in TRPM7-deficient cells cultured in regular media in conjunction with an inability of constitutive Akt signaling to support growth of TRPM7-deficient cells in regular media suggests that TRPM7-deficiency leads to failure of growth and proliferation secondary to interruption of sustained signaling along both Akt-dependent and independent growth regulatory pathways. As phosphoinositide 3-kinases are major upstream regulators of Akt and related kinases and are known to activate both Akt-dependent and -independent growth regulatory pathways (Hawkins et al., 2006), as well as exit from G₀, and progression through G₁/S and G₂/M transitions (Dangi et al., 2003; Shtivelman et al., 2002), we next examined whether provision of a heterologous sustained PI3K signal could complement the failure of TRPM7-deficient cells to grow and proliferate in regular media. To provide a sustained PI3K signal, we utilized a constitutively active myristoylated form of the p110α catalytic subunit of PI3K (myr-p110) which has been previously used successfully to identify direct downstream targets of PI3K’s and has greatly facilitated the analysis of signaling events regulated by PI3K’s (Auger et al., 2000; Didichenko et al., 1996; Klippel et al., 1996). TRPM7-deficient cell lines were generated with doxycycline inducible expression of myr-p110, and a clone with undetectable basal and easily detectable induced expression was chosen for further analysis (Figure 4A, left panel). Induction of myr-p110 in this cell line generated a clearly detectable enhancement of Akt activation in both regular and growth supporting media, demonstrating that the induced myr-p110 was expressed at a sufficient level and was appropriately activating downstream effectors (Figure 4A, right panel). When myr-p110 expressing TRPM7-deficient cells were placed in longer term culture in regular media, they exhibited obvious differences in growth and proliferation relative to uninduced cells (Figure 4B). A comparative cell cycle analysis of myr-p110 expressing cells at approximately 10 days post induction showed a clear shift of cells into the S phase relative to those which were not induced, although not to the degree observed in myr-p110 TRPM7-deficient cells cultured in growth supporting media with or without doxycycline (Figure 4B, left panel). Myr-p110 expression at 10 days post induction also allowed TRPM7-deficient cells cultured in regular media to retain their normal size (equivalent to that of WT DT40 cells), and increased the size of TRPM7-deficient cells cultured in growth supporting Mg²⁺ supplemented media to supranormal levels (Figure 4B, right panel), directly demonstrating the capacity of the myr-p110 to restore failed growth signaling in TRPM7-deficient cells. Furthermore, the proliferation rate of TRPM7-deficient cells expressing myr-p110 was observed to increase over time in culture, and a subsequent analysis of these cells at 18 days post-induction showed that, relative to 48 hours induction, induced cells had accumulated a much higher level of myr-p110 (Figure 4C, left panel), and exhibited a correspondingly greater shift of cells out of G₀/G₁ into S and sustained proliferation (Figure 4C, right panel). These results suggest that myr-p110 provides a dose-dependent enhancement of growth signaling in TRPM7-deficient cells that are selected for by culture in regular media. This possibility was directly evaluated by analysis of Akt and S6K1 phosphorylation in TRPM7-deficient cells cultured in regular media after >18 days in culture, which demonstrated that the enhanced myr-p110 expression at this time point sustained full activation of Akt (Figure 4D, left panel) and nearly full activation of S6K1 (Figure 4D, right panel). Treatment of cells with wortmannin (Figure 4D), a PI3K inhibitor, resulted in significant reduction in the phosphorylation levels of Akt and S6K1, demonstrating their continued dependence on phosphoinositide 3-kinase signals.

Figure 4. Heterologous sustained PI3K signaling complements growth and growth regulatory signaling failures of TRPM7-deficient cells in regular media.

A – Sustained PI3K signaling is required for sustained Akt activation in TRPM7-deficient cells expressing doxycycline-inducible myr-p110α. Left panel: Western immunoblotting analysis of a TRPM7.KO clone with inducible expression of a 6X-His tagged myristoylated form of p110α. Right panel: Western immunoblotting analysis of Akt activation in the presence and absence of 48 hours doxycycline induction, followed by 5 hours culture in regular media, and either continued culture in that media (−) or addition of 15 mM supplemental Mg²⁺ (+) for the indicated times. Myr-p110α expression produces enhanced sustained activation of the PI3K effector Akt, albeit to levels slightly less than those observed in Mg²⁺-supplemented media (+).

B – Heterologous sustained PI3K signals complements the cell cycle arrest and cell size reduction which occur when TRPM7-deficient cells are transitioned to regular media. Left panel: DAPI DNA content analysis of TRPM7-deficient cells in the indicated conditions, demonstrating that induction of myr-p110α expression allows exit of TRPM7-deficient cells from G₀/G₁ arrest. Right panel: Flow cytometric analysis of the cell size of WT, myr-p110 expressing TRPM7-deficient cells cultured in regular media (−), or myr-p110 expressing cells cultured in Mg²⁺ supplemented media (+). While TRPM7-deficient cells normally are substantially smaller than WT DT40, myr-p110 expression restores their size to the equivalent of DT40 cells when cultured in regular media, and to supranormal size when cultured with supplemental Mg²⁺.

C – Prolonged growth of TRPM7-deficient cells expressing constitutively active mp110-PI3K results in selection of cells with enhanced mp110-PI3K expression and cell cycle complementation. Left panel: Western immunoblotting analysis of myr-p110α expression in TRPM7-deficient cells induced in Mg²⁺ supplemented media (48 hours induction), and after 18 days of induction and culture in regular media followed by the indicated number of days without doxycycline. Right panel: DAPI DNA content analysis of cell cycle distribution of myr-p110 expressing cells cultured in regular media (−) with doxycycline added/withdrawn for the indicated periods of time.

D – mp110-PI3K expression provides sustained growth regulatory signaling to TRPM7-deficient cells. Left panel: Western immunoblotting analysis of Akt Ser473 phosphorylation after 3 weeks of induction of myr-p110α expression. myr-p110 expressing TRPM7-deficient cells were cultured in regular media for 5 hours, and then continued in that media (−) or supplemental Mg²⁺ was added (+). High level of myr-p110α expression in these cells produces fully sustained activation of the PI3K effector Akt which is wortmannin sensitive. Right panel: Western immunoblotting analysis of S6K1 Thr389 phosphorylation after 3 weeks of induction of myr-p110. Myr-p110 expressing TRPM7-deficient cells were cultured in regular media for 5 hours, and then continued in that media (−) or supplemental Mg²⁺ was added (+). High level of myr-p110α expression in these cells produces nearly fully sustained activation of the S6K1 translational regulator, which is wortmannin sensitive.

All data are representative of at least two independent sets of experiments.

Discussion

We have used cell physiological, biochemical, and genetic complementation assays to gain insight into the unique Mg²⁺-dependent proliferative arrest phenotype observed in TRPM7-deficient cells. We find that: 1) TRPM7-deficient cells cultured in regular media exhibit a profound failure of cell growth (addition of size and mass), and a secondary failure of proliferation; 2) TRPM7-deficient cells cultured in regular media exhibit substantially decreased signaling along the entire phosphoinositide 3-kinase/Akt/mTOR/S6K protein translation regulatory pathway; and 3) Heterologous expression of a membrane-targeted, constitutively active p110 phosphoinositide 3-kinase complements phosphoinositide 3-kinase signaling, cell growth, and proliferation of TRPM7-deficient cells grown in regular media. Overall, these results suggest that a central function of TRPM7 is to promote sustained activation of cellular phosphoinositide 3-kinase signals, thus positioning TRPM7 alongside phosphoinositide 3-kinases as a central regulator of vertebrate lymphocyte growth.

A significant question raised by our results is the nature of the mechanism through which TRPM7 is influencing phosphoinositide 3-kinase dependent signaling. The fact that supplemental Mg²⁺ is able to support growth of TRPM7-deficient cells argues that this mechanism somehow involves Mg²⁺ ions entering the cell. While evidence supporting changes in global free Mg²⁺ as a signaling phenomena has been reported (reviewed in (Takaya et al., 2000)), we could not detect global population changes in intracellular free Mg²⁺ (Figure S4). However, physiologically relevant changes in free Mg²⁺ may potentially occur within microdomains surrounding Mg²⁺ transporters. Understanding how Mg²⁺ entering via TRPM7 and other Mg²⁺ transporters influences sustained phosphoinositide 3-kinase signaling promises to yield new insights into the links between metabolism and growth regulation in lymphocytes as well as other cell types.

A final intriguing implication of our results is that there is close linkage between Mg²⁺ uptake and biosynthetic metabolism required for cell growth. Why might such a link exist? A compelling explanation lies in the different requirements for cell metabolism in quiescent vs. dividing cells. While both quiescent and dividing cells require glucose to satisfy their energy demands, quiescent cells utilize glucose to recycle existing MgADP to MgATP - both ATP and ADP exist primarily in their Mg²⁺ bound forms. As Mg²⁺ is not lost in the process of mitochondrial ATP regeneration, no new Mg²⁺ ions are required to support this form of energy production. In contrast, a cell progressing through the cell cycle utilizes glycolysis and the pentose phosphate shunt to generate substrates for addition of cell mass (reviewed in (Frauwirth and Thompson, 2004)), and in lymphocytes, rapid bursts of growth are accompanied by the largely de novo synthesis of an expanded cellular pool of ATP for apportioning between its daughters (Allison et al., 1977). As each newly synthesized ATP molecule requires a new Mg²⁺ ion be imported from the extracellular milieu, closely linking Mg²⁺ uptake and biosynthetic metabolism is an attractively simple mechanism to allow cells to switch between quiescent and proliferative metabolic states: in the former situation, Mg²⁺ uptake is limited, switching off growth pathways and focusing cellular carbon flux on (Mg²⁺) ATP regeneration; in contrast, in the latter situation, Mg²⁺ uptake is active, switching on growth pathways and focusing cellular carbon flux on synthesis of new biomolecules, including new molecules of (Mg²⁺) ATP.

Experimental Procedures

Materials

The phospho-p70 S6Kinase (S6K1) (Thr389), phospho-Akt (Ser473 and Thr308), phospho-4E-BP1 (Ser65), phospho-p44/42 MAP Kinase (Thr202/Tyr204), p70S6kinase, Akt, 4E-BP1, Erk, HA-tag and PI3 kinase p110α antibodies were obtained from Cell Signaling Technology. Rapamycin (mTOR inhibitor), wortmannin (PI3K inhibitor) and 4'-6-diamidino-2-phenylindole (DAPI) were purchased from Sigma. The cDNA encoding myr-Akt (Δ4–129; HA-tagged at C-terminus) was a kind gift from Dr. Ann Olson/Dr. Morris Birnbaum and has been previously described (Roose et al., 2007). The cDNA encoding for His–tagged myr-p110α was obtained from Addgene (Addgene plasmid 10836, (Auger et al., 2000)).

DT40 cell line construction and culture

DT40 cells were maintained in Roswell Park Memorial Institute (RPMI 1640) (Mediatech) with 10% fetal bovine serum (FBS), 1% chicken serum, 10U/ml penicillin/streptomycin, 2mM glutamine and 50µg/ml blasticidin. Construction of the TRPM7-KO cell line in a DT40 parental cell line expressing the tetracycline transcriptional suppressor using pCDNA6/TR (Invitrogen) and complementation of its growth with MgCl₂ has been previously described (Schmitz et al., 2003).

Cell size and cell cycle analysis

DNA content and cell cycle analysis was carried out after fixation of cells and staining them with DAPI. Cells were acquired on LSRII flow cytometer (BD Biosciences) and analyzed by Flowjo (Ashland, Oregon). Briefly, 1–2×10⁶ cells were spun and washed once with PBS. Cells were then fixed by adding 1ml of ice-cold 70% ethanol and storing them at 4°C for at least 1 hour, following which the cells were washed twice with PBS. DAPI (1µg/ml) dissolved in PBS, 0.1% BSA and 0.1% TritonX-100 was added to the cells and mixed well. Cells were kept at 4°C for 30 min and analyzed on the LSRII flow cytometer (BD Biosciences) with ultraviolet excitation and DAPI emission collected at >450nm. Percentages of cells existing within the various phases of the cell cycle were calculated by gating on G₀, G₁, S, and G₂/M cell populations visualized using the cell cycle platform in FlowJo. Forward scatter (FSC) was used to quantify the cell size.

Establishment of an optimal time frame for phosphorylation analysis

For analysis of cells under growth-supporting and growth-arresting conditions, DT40 WT cells (used as a control) were washed twice with PBS and resuspended in fresh RPMI media for 2 hours at 37°C prior to starting the experiment so as to negate any effects mediated by exposure to fresh nutrients in the media. The cells were then counted and divided into three groups:(a) cells in regular RPMI without any supplemental Mg²⁺, (b) cells in regular RPMI with 15mM supplemental Mg²⁺ and (c) cells in regular RPMI with 15mM supplemental Mg²⁺ and 100nM rapamycin/insulin/wortmannin depending on the experiment being conducted. TRPM7-KO cells, which grow only in the presence of 10–15mM supplemental Mg²⁺, were treated in a slightly different manner. While investigating for an optimal time frame, TRPM7-KO cells were washed twice with PBS and transferred to fresh RPMI either with or without supplemental Mg²⁺ and analyzed for S6K1 phosphorylation from 1–5 hours, respectively. A sizeable increase in S6K1 phosphorylation was observed as early as 0.5–1 hour after transfer to regular as well as growth-supporting media (containing supplemental Mg²⁺) as can be seen in Figure S1, left panel. To establish an approximate time point of down-regulation of S6K1 phosphorylation, TRPM7-KO cells were washed twice with PBS and transferred to fresh media with 15mM [Mg²⁺]_ex for 2 hours. Subsequently, the cells were washed with PBS twice and transferred to fresh RPMI either with or without supplemental Mg²⁺ and analyzed for S6K1 phosphorylation till 8 hours. A detectable decrease in phosphorylation of S6K1 was observed starting from 3 hours post-transfer of TRPM7-deficient cells from growth supporting to regular media, and was substantially down-regulated by 5 hours post-transfer, indicating that TRPM7-KO cells are able to sense and respond to Mg²⁺ limitation within a time frame of 5 hours with minimal side effects (Figure 2A, upper left panel). For further experiments, TRPM7-KO cells were washed twice with PBS and resuspended in fresh RPMI media without supplemental Mg²⁺ for 5 hours at 37°C prior to starting the experiment. The cells were counted and divided into three groups with a similar approach as for WT DT40 cells, harvested at various time points and stored at −80°C till further analysis after washing with PBS.

Whole cell lysate preparation, Electrophoresis and Western blotting

After treatments, cells were washed once with PBS and whole cell lysates were prepared by lysing the cells in ice-cold lysis buffer [20mM Tris-HCl, pH 7.4, 120mM NaCl, 20mM NaF, 1mM EDTA, 6mM EGTA, 20mM β-glycerophosphate, 0.5mM DTT, 1mM sodium vanadate (Na₃VO₄), 1% NP-40] and complete mini protease inhibitor cocktail without EDTA (Roche), used according to manufacturer’s instructions. The lysates were rotated for 45 min at 4°C and cell debris was removed by centrifugation at 14,000 rpm for 15 min at 4°C. Protein concentration of the lysates was determined either by Bradford protein assay (Bio-Rad) or Bicinchoninic acid (BCA; Pierce) using the manufacturer’s specifications. Gels were loaded with 50–60µg of protein/sample per lane. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out according to the method of Laemmli. Aliquots of the supernatant were separated on SDS-PAGE gels (15% for 4E-BP1 and 10% for Akt, S6K1, Erk and 8% for p110α) and were analyzed by immunoblotting (IB). The proteins were transferred to 0.45µm pore size PVDF membranes (Polyvinylidene fluoride; Millipore) in transfer buffer (39mM glycine, 48mM tris base and 20% methanol) for 3–4 amp hours at 4°C. Membranes were blocked in 5% blocking buffer (5% w/v nonfat dry milk in TBS-0.1% Tween-20) for 1 hour at room temperature. Primary antibody incubations were done overnight at the dilutions specified by the vendor. Incubations with the secondary antibody were performed with peroxidase coupled anti-rabbit/anti-mouse immunoglobulin in 5% blocking buffer and the bound antibody was detected by the ECL chemiluminescence detection system (Amersham). For p-S6K1 quantitation, Western blots from three independent experiments were scanned, and the specific p-S6K1 bands were densitometrically quantitated with Adobe Photoshop CS software (Adobe Systems, San Jose, CA). Band densities were corrected for background using a reference area close to each band, and p-S6K1 band intensities were normalized for loading by normalization to the total S6K1 band intensity obtained by stripping and reprobing the same blot. P values for the fold change were determined using a one-tailed Student's t-test assuming equal variances.

Myristoylated Akt (myr-Akt) and myr-p110 PI3K cloning & expression

Myr-Akt comprises murine Akt-1 with the src myristoylation sequence fused to the N-terminus and hemagglutinin (HA)-tag at the C-terminus. Myr-Akt was cloned into pcDNA5/TO vector (Invitrogen, T-rex system), which provides a tetracycline-controlled expression from a cytomegalovirus (CMV) immediate early promoter and its expression was induced by overnight treatment with 1µg/ml doxycycline. The pcDNA5/TO-myrAkt-HA construct was transfected by electroporation into TRPM7-KO DT40 cells expressing the tet repressor protein (via previous stable transfection with pCDNA6/TR). Electroporation was carried out at 550V/25µF in 4mm electroporation cuvettes (BioRad GenePulser X cell) and the cuvettes were placed on ice for 5 minutes before and after pulsing. Cells were cultured in RPMI-1640 media supplemented with 10% FBS, 1% chicken serum, 10U/ml penicillin/streptomycin and 2mM glutamine overnight, prior to plating them into 96 well plates at various dilutions, under antibiotic selection along with addition of 15mM Mg²⁺ for the TRPM7-KO cells. The resistant clones were selected with hygromycin (2mg/ml; Calbiochem) for myr-Akt-HA transfection and were analyzed for doxycycline-induced protein expression by western blotting with the anti-HA antibody. Its phosphorylation status was further confirmed with the phospho-Akt S473 antibody. The myr-p110α-His expression construct was made by cloning myr-p110α-His into pcDNA5/TO followed by transfection into TRPM7-KO cells. The resistant clones were selected and analyzed for protein expression by immunoprecipitation with Ni²⁺-NTA agarose according to a specified protocol. Further PI3K modulation studies were carried out in presence/absence of 15mM Mg²⁺.

Measurement of intracellular Mg²⁺

DT40 WT and TRPM7-KO cells were harvested at the indicated time points, washed twice with PBS and resuspended in fresh RPMI or were kept in the original media in which they had been growing. Cells were then labeled with mag-indo-1 (Molecular Probes, Eugene, OR, U.S.A) dissolved in dimethyl sulphoxide (DMSO, Fisher) and incubated for 35 mins at 37°C under 5% CO₂. After washing twice with PBS, cells were resuspended either in HBSS without Ca²⁺ and Mg²⁺ and 2% FCS or in regular RPMI and blue/violet fluorescence data were collected on the LSR II flow cytometer at room temperature. Data were subsequently analyzed off-line using Flowjo (Ashland, OR, U.S.A).