A TRPM7 variant shows altered sensitivity to magnesium that may contribute to the pathogenesis of two Guamanian neurodegenerative disorders

Abstract

Guamanian amyotrophic lateral sclerosis (ALS-G) and parkinsonism dementia (PD-G) have been epidemiologically linked to an environment severely deficient in calcium (Ca2+) and magnesium (Mg2+). Transient receptor potential melastatin 7 (TRPM7) is a bifunctional protein containing both channel and kinase domains that has been proposed to be involved in the homeostatic regulation of intracellular Ca2+, Mg2+, and trace metal ion concentration. There is evidence that TRPM7 is constitutively active and that the number of available channels is dependent on intracellular free Mg2+ levels. We found a TRPM7 variant in a subset of ALS-G and PD-G patients that produces a protein with a missense mutation, T1482I. Recombinant T1482I TRPM7 exhibits the same kinase catalytic activity as WT TRPM7. However, heterologously expressed T1482I TRPM7 produces functional channels that show an increased sensitivity to inhibition by intracellular Mg2+. Because the incidence of ALS-G and PD-G has been associated with prolonged exposure to an environment severely deficient in Ca2+ and Mg2+, we propose that this variant TRPM7 allele confers a susceptibility genotype in such an environment. This study represents an initial attempt to address the important issue of gene-environment interactions in the etiology of these diseases.

Fig. 1.

Chromatograms of ALS-G, PD-G, and control sequences. A heterozygous C-to-T transition (arrowhead) was found in ALS-G and PD-G samples but not in controls. This transition causes an exon mutation where Thr is replaced with Ile (box). Sequences are in the reverse (minus-strand) orientation. Numbers in parentheses are numeric sample codes used in the blind study.

Fig. 2.

Thr-1482 is in an α-helix between the channel and kinase domains and is evolutionarily conserved. (a) Schematic representation of TRPM7. Residue 1482 is found between the coiled-coil and kinase domains. (b) Secondary structure analysis predicts Thr-1482 as part of an α-helix, whereas Ile-1482 becomes part of a coil. (c) Alignment of the appropriate region of TRPM7 from different species shows evolutionary conservation of Thr-1482 (boxed), except in the mouse, where serine (Ser) is present instead. National Center for Biotechnology Information (GenBank, Entrez Protein) accession numbers are shown in the right. Orangutan and Xenopus sequences were translated from ESTs CR769569 and CA973711. Sequences were aligned by using clustalw (http://www.ch.embnet.org/software/clustalw.html).

Fig. 3.

Assessment of inducible expression and immunolocalization of expressed WT and T1482I. (a) RT-PCR of inducible HEK-293 cells stably transfected with WT and T1482I in the presence and absence of the inducer, DOX. The mutant clone selected exhibits channel expression levels that closely match WT expression after induction. Faint bands detected in the absence of DOX represent low-level expression of endogenous TRPM7. (b) Sequence chromatograms of the RT-PCR products from induced cells in a confirm the genotype of the expressed channels (arrowheads). Primers for the plus strand were used for the sequencing reactions. (c) Anti-HA immunofluorescent staining of HEK-293 cells induced to express WT and T1482I channels. The same pattern of punctate membrane and cytoplasmic staining indicates that the mutation does not alter channel trafficking and localization.

Fig. 4.

Whole-cell currents and sensitivity to internal Mg²⁺. (a) Representative whole-cell currents elicited by a 50-ms voltage ramp from -100 mV to + 100 mV in cells overexpressing WT and T1482I perfused with nominal 0 Mg²⁺ pipette solution. Data traces taken at whole-cell break-in (t = 0), 5 min, and 10 min into the experiment are shown. All recordings were measured in cells that have been exposed to DOX for 24–36 h. Small currents with the pronounced outward rectification characteristic for TRPM7 can be detected at break-in. The currents developed as cells were dialyzed, reaching near-maximal levels within 3 min. These results show that T1482I channels are functional. (b) Time course of development of WT and T1482I currents in the presence of various concentrations of internal Mg²⁺. Plots represent average current densities measured at + 80 mV for each Mg²⁺ concentration used: WT, 0 mM Mg²⁺, n = 8; 0.5 mM Mg²⁺, n = 6; 1 mM Mg²⁺, n = 7; 2 mM Mg²⁺, n = 10; 4 mM Mg²⁺, n = 6; and 6 mM Mg²⁺, n = 6; T1482I, 0 mM Mg²⁺, n = 15; 0.5 mM Mg²⁺, n = 5; 1 mM Mg²⁺, n = 12; 2 mM Mg²⁺, n = 5; 4 mM Mg²⁺, n = 5; and 6 mM Mg²⁺, n = 4. Averaged traces were normalized to the break-in value at 0 Mg²⁺. (c) T1482I channels are more susceptible to inhibition by internal Mg²⁺. Current densities measured at -80 mV and +80 mV are plotted as a function of pipette Mg²⁺ concentration. The data points were obtained by subtracting the break-in value measured in each cell from the maximum current density attained during the first 400 s of the experiment and averaging the results for each Mg²⁺ concentration used (±SEM).

Fig. 5.

Comparison of kinase catalytic activity of WT and T1482I mutant TRPM7 and phosphoamino acid analysis. (a) Purified recombinant human TRPM7 (amino acids 1403–1864) and T1482I mutant were incubated with [γ-³³P]ATP in the kinase reaction mixture as described in Methods. The samples were analyzed by SDS/PAGE and autoradiography. (b) Phosphoamino acid analysis of autophosphorylated WT TRPM7 and T1482I mutant (amino acids 1403–1864). Phosphoamino acid analysis was performed by hydrolysis of phosphoproteins with HCl, separation of amino acids by using 2D electrophoresis on TLC plates, and autoradiography (see Methods). Quantitation of the amount of radioactivity incorporated into phosphoserine and phosphothreonine was performed by using a PhosphorImager.