Abstract: Transient receptor potential melastatin 7 (TRPM7) is a widely expressed, constituatively active, voltage independent ion channel known to be permeable to most divalent and monovalent cations. TRPM7 may play an important role in Ca, Mgr, and trace metal homeostasis. Our study examines the behavior of TRPM7 under physiologically relevant conditions. Confocal laser microscopy and the indicator dye fluo-3/AM allowed for visualization of intracellular Ca“ accumulation in control and TRPM7 overexpressing HEK-293 cells when exposed to external solutions with varying concentrations of Ca“ and Mg“. When compared with control cells, TRPM7 overexpressing cells show greater influx and efflux of Ca“ in response to changes in external Ca“, indicating that TRPM7 functions as a calcium channel under physiologically appropriate conditions. Our results also indicate that Mg“ interferes with the movement of Ca“ through the TRPM7 channel. The data further suggest that the percentage of TRPM7 channels blocked by Mg“ is proportional to the extracellular Mg concentration. This implies a possible mechanism for ion permeation through TRPM7 that involves ion-ion interactions within the channel protein and implicates TRPM7 channels in Ca“ and Mg homeostasis. Introduction: Transient receptor potential (TRP) channels were first identified in a drosophila mutant characterized by photoreceptors that produce abnormally short receptor potentials in response to light (Montell 2002). The melastatin sub-family of TRP channels is common in many mammalian cells including those found in the lymph nodes, thymus, bone marrow, and brain (Wolf 2004). TRPM7 knockouts die unless given supplemental magnesium, indicating that the ion channel is essential for normal cell functions (Schmitz et al. 2003). Overexpression of the ion channel can also cause cell death possibly due to apoptosis triggered by the influx of Ca through TRPM7 (Nadler 2001). TRPM7 may be involved in Ca“, Mg“, and trace metal homeostasis. TRPM7 is regulated by intracellular Mg- and Mg“-ATP, and the ion channel conducts several essential and non-essential trace metals, including Zn2, Ni2, Ba2, Co*, Mn2, Sr2, and Cd“ (Monteilh-Zoller et al. 2003). Imaging studies using the indicator dye fura 2/AM have suggested that TRPM7 provides a pathway for Ca“ entry into the cell (Monteilh- Zoller et al. 2003). Patch clamp and imaging studies have demonstrated Mn* influx in TRPM7 expressing cells in the presence of Ca and Mg“, suggesting that Mn2 competes with Ca“ and Mg- for entry through TRPM7 (Thompson and Hermosura, unpublished). Homeostatic regulation of Ca“, Mg-, and trace metals is required for normal cell function. Mg¬ is a cofactor for many enzymes involved in cellular metabolism. Mg¬ deficiency has been implicated in Chronic Fatigue Syndrome, and a variety of cardiovascular and neurological problems, including stroke and hemorrhage (e.g. Altura et al. 1998; Amighi et al. 2004; Van den Bergh et al. 2003). Some trace metals also serve as cofactors for enzymes, but abnormal levels of trace metals have been correlated with Alzheimer’s and Parkinson’s disease and related syndromes. Aluminum accumulation in the amyloid plaques associated with Alzheimer’s disease has been assumed to play a role in the etiology of the disease, although this assumption has recently been called into question (Kasa et al. 1995). Öther research has shown that copper and zinc bind to amyloid beta, which may explain accumulation of these trace metals in amyloid plaques (Atwood 2000). Parkinson’s related cell death in the substantial nigra (SN) has been correlated with high levels of free iron, low levels of the iron-binding protein ferritin, high levels of zinc, and low levels of copper in SN tissue (Dexter et al. 1992). African green monkeys with unilateral MPTP-induced parkinsonism showed high concentrations of Fe“ and Fe" in the substantia nigra compacta on the MPTP-lesioned side, leading researchers to conclude that high concentrations of unbound iron may be linked to cell death in the SN (Temlett et al. 1994). However, pathology associated with trace metal accumulation may be better explained by high levels of trace metals coupled with low levels of Ca and Mg' During an epidemic of Parkinson’s-ALS symptoms in Guam in the 1950s, many patients developed severe dementia and most died within five years of the onset of symptoms; these Parkinson’s ALS clusters occurred in regions of Guam with high levels of Aluminum and Manganese coupled with low levels of calcium and magnesium in the soil and river water (qtd. in Yasui and Garruto 1995). Laboratory studies have shown that rats fed a diet low in Ca“ and Mg- with or without high aluminum showed increased aluminum deposition in central nervous system tissue (Yasui and Garruto 1995). Pathology related to disturbance in trace metal and earth metal homeostasis may become more prevalent in the future because industrial use of trace metals is increasing. For example, the fungicides Maneb and Mancozeb and the gasoline additive MMT (antiknock 33x) all contain manganese. Better understanding of the structure and function of TRPM7 may help in understanding Ca and Mg- homeostasis and the pathology that results from disturbance of metal homeostasis. Electrophysiological studies have concluded that TRPM7 is a nonselective cation channel permeable to most mono and divalent cations (e.g. Kerschbaum et al. 2003), however when divalent cations are present in the external solution TRPM7 only passes divalent cations. We propose that TRPM7 may be a calcium channel that can be forced to pass monovalent cations under the artificial conditions required for voltage clamp studies. While voltage clamp studies have provided valuable information that has ultimately led to the current four barrier, three well Eyring rate theory model of the channel pore, little is known about the channel’s behavior under more physiologically relevant conditions. Our study examines the behavior of TRPM7 in the presence of physiologically appropriate levels of Ca and Mg. Our results indicate that TRPM7 functions as a calcium channel in solutions that approximate normal physiological conditions. We further demonstrate that magnesium blocks Ca“ flux through TRPM7 in a dose dependant manner. Materials and Methods: Cell Culture Up-regulation of TRPM7 expression was achieved using the A10 cell line generously provided by A. Sharenberg. A10 cells were created from human embryonic kidney (HEK-293) cells stably transfected with murine TRPM7/pCDNA under the control of a tetracycline-inducible promotor. Cells were cultured in Dulbecco’s modified eagle’s medium (DMEM) with 10% Fetal Bovine Serum (FBS), 0.1% penicillin, 0.1% streptomycin, 5ug/ml blastocidin, and 0.4 mg/ml zeocin. Cells were kept in a 30° C incubator with 5% CO2 and passaged every 2-4 days when barely confluent. Äfter each passage, cells were plated onto glass cover slips that had been coated with high molecular weight poly-lysine and rinsed with nanopure water to increase cell adhesion. 24-48 hours after plating, half the cells were induced with lug/ml tetracycline (TET) so that they would begin to up-regulate expression of TRPM7. Sister cells from the same plating were left uninduced as controls. Control and TET induced cells received the same treatment in every other respect. TET induced and control cells show morphological differences that aided in differentiating between the two groups (fig. 1). Solutions Cells on the glass coverslips were transferred from the tissue culture medium to a buffered saline solution of the following composition (mM): 140 NaCl, 2.8 KCl, 10 Hepes-NaÖH, 2 CaClz. Experimental solutions containing different concentrations of Ca“ and Mg“ were made using the concentrations listed in table 1. Solutions containing 0.2, 0.02, and 0.002 mM Mg“ were made from a OmM Ca“, OmM Mg stock solution containing 2mM metal chelator, EGTA (table 1). The amount of MgClz added to the stock solution was adjusted to account for EGTA binding to free Mg“. Calculations were made using MaxChelator (Patton et al. 2004). All solutions were titrated with IN or O.IN NaÖH to bring the pH to 7.2. Glucose (approximately.02g/5Oml) was added to all solutions a few hours before use. Imaging The Ca“ indicator dye fluo3/AM (Molecular Probes) increases in fluorescence when bound to intracellular Ca“. Cells in 2mM Ca“ buffered saline solution were loaded with fluo-3 (2.5uL/ml external solution) and gently agitated for 40 min. Cells were then washed with fresh 2mM Ca“ saline at least 20 min before the start of imaging. The concentration of Ca“ inside the cell was visualized using an Olympus Fluoview 300 laser scanning microscope with a UPLFL 20x lens (fig. 2). For viewing under the microscope, coverslips were transferred individually to a plexiglass perfusion chamber that held Icc of saline solution. A perfusion system consisting of a lcc syringe with attached microtubing delivered the experimental solutions to the plexiglass chamber and excess solution was drawn off using a vacuum pump. Solution changes were made over the course of 60 seconds. Data was collected every 4 seconds for a total 240 seconds. Data Analysis Individual cells were defined as regions of interest (ROIs) using Fluoview Physiology 4.1 (Tiempo Physiology). Isolated cells or cells with few confluent neighbors were chosen in order to minimize complications due to cell-cell interactions. Typically ten regions of interest were defined for each trial. Measurements of fluorescence intensity over time in each ROI were made using Fluoview Physiology. We used Igor Pro version 4.01 (Wave Metrics) to create graphs of fluorescence versus time for individual ROIs and for the average of all ROIs in each trial. In order to compare graphs from different trials and to account for differences in dye loading across trials, we calculated a baseline initial fluorescence (Fo) by averaging all the data points collected before perfusion with a new solution (baseline=O to 20 seconds). We then normalized the wave by dividing fluorescence at each time point (F) by the initial fluorescence (Fo) to obtain values of AF/ Fo. Trials in which the same treatment was applied (e.g. removal or addition of extracellular Ca“) were averaged using Igor Pro. When applicable, the results are reported as mean value +/- one standard error. The percent block graphs were designed using the fluorescence intensity of cells bathed in 0 mM Ca“ saline solution as the hypothetical maximum block of TRPM7. Percent block was then calculated by comparing fluorescence intensity at the end of each trial (Ffinal) and plotted on a semilog scale to accommodate the range of Mg? concentrations studied. Results: Response to Changes in Extracellular [Cat 1: For TET induced TRPM7 overexpressing cells in 2mM Ca2 buffered saline, initial fluorescence (Fo) calculated from all data points before perfusion (from 0 to 20 seconds) was 0.90 +-.03 (n=12 trials). Values of AF/ Fo decreased to 0.69 +- 0.03 (n=12 trials) following perfusion with the OmM buffered saline solution (fig. 3). This decrease in fluorescence indicates a decrease in internal [Ca“). We can infer that the decrease in internal [Ca# is due to Ca“ efflux through TRPM7 because uninduced control cells showed a much smaller decreased in fluorescence. For control cells, in 2mM Ca“ buffered saline, initial fluorescence (Fo) was 0.99 + 0.01 (n=12 trials). Values of AF/ Fo decreased to 0.96 +-0.01 (n=12 trials) following perfusion with OmM Ca“ buffered saline (fig. 3). TET induced cells also show a greater change in internal [Ca“ in response to addition of external calcium when compared with uninduced control cells. For TET induced cells in OmM Ca“ buffered saline, initial fluorescence (Fo) was 1.00 +-0.00 (n—9 trials). Values of AF/ Fo increased to 1.42 +- SE= 0.06 (n-9 trials) following perfusion with the 2mM buffered saline solution (fig. 4). This increase in fluorescence indicates an increase in internal [Ca. We can infer that this increase in internal [Ca? is due to Ca“ influx through TRPM7 because un-induced control cells did not show a similar increase. For control cells, in OmM Ca“ buffered saline, initial fluorescence (Fo) was 1.00 +- 0.01 (n=12 trials). Values of AF/ Fo did not change 1.01 +-0.01 (n=12 trials) following perfusion with OmM Ca“ buffered saline (fig. 4). Response to the Addition of Mg For TET induced TRPM7 overexpressing cells in 2mM Ca2/OmM Mg buffered saline, fluorescence values of AF/ Fo increased .03 units (n=10 cells) following perfusion with the 2mM Ca“7.002mM Mg“ buffered saline solution. The same cells showed a decrease in AF/ Fo of.15 units (n=10 cells) following perfusion with 2mM Ca/02mM Mg buffered saline, .17 units (n=10 cells) following perfusion with 2mM Ca2/.2mM Mg“ buffered saline and .28 units following perfusion with 2mM Ca?/2mM Mg buffered saline solution (fig. 5a). This decrease in fluorescence indicates a decrease in internal [Ca“]. We can infer that the decrease in internal [Ca* is due to Ca* efflux through TRPM7 because the uninduced control cells showed a much smaller decreased in fluorescence. For control cells, in 2mM Ca“ buffered saline, values of AF/ Fo increased 1 units (n=10 cells) and .05 units (n=10 cells) following perfusion with 2mM Ca“7.002mM and 2mM Ca“7.02 mM Mg buffered saline, respectively and decreased 02 units (n=10 cells) and .06 units (n=10 cells) following perfusion with 2mM Ca“7.2mM and 2mM Ca“/2mM Mg buffered saline, respectively (fig. 5b) Response to the Removal of Mg For TET induced TRPM7 overexpressing cells in 2mM Ca21/.002mM Mg? buffered saline, fluorescence values of AF/ Fo increased.07 units (n-10 cells) following perfusion with 2mM Ca“/OmM Mg- buffered saline solution. The same cells showed an increase in AF/Fo of.2 units (n=10 cells), .25 units (n=10 cells) and .22 units (n=10 cells) when buffered in 2mM Ca“7.02 mM Mg“, 2mM Ca/.2mM Mg* and 2mM Ca21/2mM Mg“ buffered saline, respectively, following perfusion with 2mM Ca/OmM Mg buffered saline solution (fig. 6a). This increase in fluorescence indicates a increase in internal [Ca“. We can infer that the increase in internal [Ca2 is due to Ca2 influx through TRPM7 because the uninduced control cells do not show an much increase in fluorescence. The control cells in this experiment produced values of AF/ Fo that did not deviate beyond +-.03 (n—4 trials) (fig. 6b). Response to the Removal of Ca“ from Solutions of Varied Mg Molarity For TET induced TRPM7 overexpressing cells in 2mM Ca?/.002mM Mg? buffered saline, fluorescence values of AF/ Fo decreased .40 units (n-10 cells) following perfusion with 0 mM Ca“7.002mM Mg buffered saline solution. The same cells showed a decrease in AF/ Fo of 21 units (n=10 cells), .23 units (n=10 cells) and .08 units (n=10 cells) when buffered in 2mM Ca“/.02mM Mg“, 2mM Ca21/.2mM Mg* and 2mM Ca“ /2mM Mg buffered saline, respectively, following perfusion with solutions containing no Ca“ (fig. 7a). This decrease in fluorescence indicates a decrease in internal [Ca“. We can infer that the decrease in internal [Ca2 is due to Ca2 efflux through TRPM7 because the uninduced control cells do not show as much decrease in fluorescence. For control cells, values of AF/ Fo decreased.14 units (n=10 cells), ,03 units (n=10 cells), .03 units (n=10 cells) and .2 units (n=10 cells) following perfusion with 0 mM Ca“7.002mM Mg“ buffered saline, OmM Ca2/.02mM Mg buffered saline, OmM Ca7.2mM Mg buffered saline and OmM Ca2/2mM Mg2 buffered saline, respectively (fig. 7b). Discussion: This study uses change in fluorescence intensity to measure Ca“ accumulation in the cell. The assumption that Ca“ accumulation is due to Ca“ flux through TRPM7 was based on the historical reliability of the HEK 293 /AlO expression system as well as correlative evidence provided by control experiments and the experimental design (Monteilh-Zoller et al. 2003; Kerschbaum 2003; Thompson and Hermosura, unpublished). The experimental method employed throughout this study was one in which extracellular conditions could be altered during fluorescence imaging. This paradigm allowed us to address two major topics of interest: the second contingent on the first. By altering the concentration of extracellular Ca“ in the absence of Mg“ we were able to establish that TRPM7 is a viable means of Ca“ flux. Three subsequent experiments were designed to probe the effect of extracellular Mg“ on Ca flux through TRPM7. Trials involved a dose-gradient in order to best articulate the relationship between Ca“ flux and extracellular Mg“ concentration. We found that Mg- blocks Ca* flux through TRPM7 in a concentration dependent fashion. In the first experiment involving Mg“, we found that the fluorescence decreased consistently with the addition of extracellular Mg“ and concluded that Mg" does block the influx of Ca through TRPM7. Furthermore, the final fluorescence (Ffinal) was found to decrease proportionally to the increase of Mg- concentration across trials. Uninduced control cells also showed a decrease in Ffinal in response to the addition of Mg“, but on a much smaller scale. The clarity of concentration dependence suggested the usefulness of a graph depicting the relationship between Mg“ concentration and the percent blockage experienced by the TRPM7 channels (fig. 8a). The percent block graph provides some indication of the permeation mechanism involving TRPM7, Ca flux and changing extracellular Mg- concentrations. If it is assumed that the block is due to a single binding site on the channel, then Eyring rate theory would predict an exponential curve reaching a horizontal asymptote as the mechanism approaches 100% saturation. In this scenario, the Mg“ concentration that causes a half-block, or 50% channel saturation, would be considered the dissociation constant, or Kd, for that binding site. Although our study does not avail itself to definitive results in this regard, we can begin to reveal an exponential trend in the percent block graph and note a possible half-block at 100 uM Mg“. Using voltage clamp techniques to study Na’ influx through TRPM7 in the absence of intracellular Mg? possible Kd of 1 uM has been proposed (Kerschbaum et al. 2003). Although the current study differs significantly in design, the two results are not in obvious contradiction. In the second experiment of the series, the Mg“ block was removed and Ca was better able to pass through TRPM7 channels. Values of Ffinal were seen to increase in a Mg“ concentration dependent manner. A percent block graph was designed and the half- block or possible Kd was determined to be 500 uM Mg (fig. 8b). Control trials reflected a similar increasing trend on a smaller scale. Some deviation from the trend may be explained by experimental error. If this deviation appears consistently, however, it may be an indication of more complicated ion-ion interactions at the binding site. It is also notable that trials involving Ca“ influx appear to show more inconsistency between cells than trials involving Ca“ efflux (fig. 9). This inconsistency may be due to variability in the cell cycle and the mechanisms that regulate internal stores and intracellular concentrations of Ca and Mg. In the third experiment of the series, in which various extracellular concentrations of Mg“ were held constant while Ca“ was removed, results showed a sharp decrease in fluorescence. In addition, the amount of change in fluorescence decreased as the concentration of Mg“ was increased. That is to say, higher Mg- concentrations meant more blocking and less intracellular Ca“ available for efflux through TRPM7 at the start of the experiment. Again, Ca“ flux was found to be dependent on the concentration of extracellular Mg¬t. From this study alone we cannot define a permeation mechanism but we have established a clear line of inquiry for future investigation. Exemplary trials should be replaced by averages across many trials. An attempt to determine the value of half-block, or Kd of the binding site, should be made by measuring over a greater range of Mg? concentrations. A more detailed dose-dependent graph will more accurately depict the nature of ion-ion interactions at the binding site. Also, voltage clamp measurement of Ca“ current is necessary to validate our correlation of intracellular Ca accumulation and Ca“ flux through TRPM7, to better elucidate the properties of individual channel and to approximate the number of channels involved in each experiment. Conclusions: In the absence of Mg“, TRPM7 does act as a Ca“ channel. We can further conclude that external Mg- blocks Ca" influx and efflux through TRPM7 channels in a dose-dependent fashion. We have estimated that the half block or possible Kd is between 100 and 500 uM of Mg“. We can further hypothesize that it is the interaction of Mg* at the binding site(s) of the channel that is slowing or blocking movement of Ca“ through TRPM7. Most broadly, we can propose that TRPM7 play a role in Mg* and Ca homeostasis in the cell. Acknowledgements: Our warmest thanks go to Chris Patton, Christian Reilly and Peggy Lynch with special acknowledgement of Stuart Thompson for his oversight and enthusiasm. Literature Cited: Altura, B.M., A. Gebrewold, A. Zhang, B.T. Altura, R.K. Gupta. 1998. Magnesium deficiency exacerbates brain injury and stroke mortality induced by alcohol: a 31P-NMR in vivo study. Alcohol. 15(3): 181-3. Amighi, J., S. Sabeti, O. Schlager, W. Mlekusch, M. Exner, W. Lalouschek, R. Ahmadi, E. Minar, M. Schillinger. 2004. Low serum magnesium predicts neurological events in patients with advanced atherosclerosis. Stroke. 35(1):22-27. Atwood, C.S., R.C. Scarpa, X.D. Huang, R.D. Moir, W.D. Jones, D.P. Fairlie, R.E. Tanzi, A.I. Bush. 2000. Characterization of copper interactions with alzheimer amyloid beta peptides: identification of an attomolar-affinity copper binding site on amyloid beta 1-42. Journal of Neurochemisty. 75(3): 1219-1233. Dexter, D.T., P. Jenner, A.V.H. Schapira, and C.D. Marsden. 1992. Alterations in levels of iron, ferritin, and other trace metals in neurodegenerative diseases affecting the basal ganglia. Annals of Neurology. 32(supplement): S94-S100. Kasa, P., P. Szerdahelyi, H.M. Wisniewski. 1995. Lack of topographical relationship between sites of aluminum deposition and senile plaques in the Alzheimer’s disease brain. Acta Neuropathologica. 90(5): 526-531. Kerschbaum, H.H., J.A. Kozak, and M.D. Cahalan. 2003. Polyvalent cations as permeant probes of MIC and TRPM7 pores. Biophysical Journal. 84: 2293-2305. Monteilh-Zoller, M.K., M.C. Hermosura, Nadler, M.J.S., Scharenberg, A.M., Penner, R., and Fleig, A. 2003. TRPM7 provides an ion channel mechanism for cellular entry of trace metal ions. Journal of General Physiology. 121: 49-60. Montell, C., L. Birnbaumer, V. Flockerzi. 2002. The TRP channels, a remarkably functional family. Cell. 108: 595-598. Nadler, M.J., M.C. Hermosura, K. Inabe, A.L. Perraud, Q. Zhu, A.J. Stokes, T. Kurosaki, J.P. Kinet, R. Penner, A.M. Scharenberg, and A. Fleig. 2001. LTRPC7 is a Mg¬ ATP-regulated divalent cation channel required for cell viability. Nature. 411:590-595. Patton, C., S. Thompson, D. Epel. 2004. Some precautions in using chelators to buffer metals in biological solutions. Cell Calcium. 35: 427-31 Schmitz, C., A.L. Perraud, C.O. Johnson, K. Inabe, M.K. Smith, R. Penner, T. Kurosaki, A. Fleig, A.M. Sharenberg. 2003. Regulation of vertebrate cellular Mg“ homeostasis by TRPM7. Cell. 114: 191-200. Temlett, J.A., J.P. Landsberg, F. Watt, G.W. Grime. Increased iron in the substantia nigra compacta of the MPTP-lesioned hemiparkinsonian African green monkey: Evidence from proton microprobe elemental microanalysis. 1994. Journal of Neurochemistry. 62(1): 134-146. Thompson, S.H., and M.C. Hermosura. Unpublished. Manganese influx via TRPM7 and its regulation by calcium and magnesium. Wolf, Federica I. 2004 TRPM7: Channeling the future of cellular magnesium homeostasis? Science STKE. 233(pe23): 1-4. Van den Bergh, W.M., A. Algra, J.W. van der Sprenkel, C.A. Tulleken, G.J. Rinkel. 2003. Hypomagnesemia after aneurysmal subarachnoid hemorrhage. Neurosurgery. 52(2): 276-81. Yasui, M., and R.M. Garruto. 1995. Effects of Calcium-deficient diets on manganese deposition in the central nervous system and bones of rats. Neurotoxicology. 16(3) 511-517. Table 1. Extracellular solutions for all experiments (mM Nacl KCl Hepes Naoll 27 2 Ca 2Mg 2 Ca 12Mg 140 2 Ca 1O2Mg 140 2.8 10 2 Ca J.002Mg 140 2.8 2 Ca ONg 140 2.8 ) 0 Ca pMg 140 10 0 Ca 12Mg 2.8 140 2.8 10 O Ca JO2Mg 140 0 Ca J.002Mg 145 2.8 O CaTOMgE PH was adjusted to 7.2 with NaÖH Cach Mge .02 002 .02 .002 0 EGTA 2 Figure Legends: Fig. 1. Morphological Differences between TET induced TRPM7 overexpressing cells and uninduced control cells. Uninduced HEK-293 cells show a spindly morphology often with elongated processes. Following induction with tetracycline (TET) cells become more rounded and often contain large vacuoles. Fig. 2. TET induced TRPM7 overexpressing cells as they appear using confocal laser microscopy. Ten individual cells defined as regions of interest (ROIs). Fig. 3. Change in fluorescence over time in response to a change in external solutions from 2mM Ca buffered saline to OmM Ca“ buffered saline. Perfusion with the OmM Ca“ buffered saline occurred from 20 to 80 seconds. Red circles represent the response ET induced TRPM7 overexpressing cells. Blue triangles represent the response of uninduced control cells. Fig. 4. Change in fluorescence over time in response to a change in external solutions from OmM Ca“ buffered saline to 2mM Ca“ buffered saline. Perfusion with the 2mM Ca“ buffered saline occurred from 20 to 80 seconds. Red circles represent the response of TET induced TRPM7 overexpressing cells. Blue triangles represent the response of uninduced control cells. Fig. 5. Change in fluorescence over time in response to a change in external solutions, from OmM Mg“ to one of four concentrations of Mg*. Ca concentration is maintained at 2mM Ca“. (a) TET induced cells overexpressing TRPM7 (b) Control cells Fig. 6. Change in fluorescence over time in response to a change in external solutions containing one of four concentrations of Mg to external solutions containing no Mg Ca“ concentration is maintained at 2mM Ca“. (a) TET induced cells overexpressing TRPM7 (b) Control cells Fig. 7. Change in fluorescence over time in response to the removal of Ca- from external solution. In each trial, Mg“ concentration is maintained at one of four concentrations. (a) TET induced cells overexpressing TRPM7 (b) Control cells Fig. 8. Percent block of TRPM7 as a function of Mg concentration. (a) % block calculated from experiment in which the concentration of Mg“ is increased while external Ca“ concentration is maintained. (b) % block calculated from experiment in which the concentration of Mg“ is decreased while external Ca“ concentration is maintained. Fig. 9. Individual cells show considerable variation in their response to a change in external solution from OmM Ca“ to 2mM Ca Fig. 1 Control Cells TET Induced ) Fig. 2 Fig. 3 g r eengeen 1.0 0.90 - 0.8: — TET Induced 0.80 - - Contro ae g ge e 0.70 - 100 150 200 Time (seconds) Fig. 1.4 - 1.3- + 12- 1.1 - 1.0- eeee Contro T Induced 100 150 200 Time (seconds) Fig. 5 A 1.2 TET Induced Cells 1.1- 0 mM to.002 mM 10- 0.9- 0 mM to.02 mM 0.8- 0 mM to.2 mM 07- 0 mM to 2 mM 06- O mM Mg2t, A Ca? 05+ 100 50 150 200 Time (s) Control Cells 0 mM to .002 mM W 0 mM to.02 mM 1o Ae 0 mM to.2 mM 0 mM to 2 mM 07- 06+ 150 200 Time (s) Fig. A 1.00 .002 mM Mg2t 096 0.90 - 02 mM Mg2 2 mM Mg2 085- 30.80 o D.75- 2 mM Mg2 0.70- OmM Ca TET Induced Cells 0.65- 50 100 150 200 Time (s) B 2 mM Mg .002 mM Mg2 1.0- Wa B 02 mM Mg 0.9- 2 mM Mg? 0.8 0.7- Control Cells 150 50 100 200 Time (s) Fig. 7 A 1.4- 113- 1.2 511 10 1.4 1.3- 1.2 11- 1.0 .002 mMMg? 02 mM Mg2t 2 mMMg?. H 2 mM Mg V W 100 150 200 Time (s) 2 mM Mg? 002 mM Mg2 02 mMNg K V W 2 mMMgt 50 100 150 200 Time (s) Fig. 8 A 60 50 - 40- 30- § 20- 10- 0-1 2.3 4 5678 2 3 45678 0.01 Mg2 concentration (mM) B 60 - 50- 40- 30- 20- 10- 2 3 45678 2 3 45678 0.01 Mg concentration (mM) 2 3 45678 2 3 45678 2 2 Fig. 350- ndividual Cells 550592 0O to 20 lo 3000 2500 - V 2000 1500- - 1000 - a- — 100 50 150 200 Time (seconds)