Ca2+-dependent induction of TRPM2 currents in hippocampal neurons

Abstract

TRPM2 is a Ca(2+)-permeable member of the transient receptor potential melastatin family of cation channels whose activation by reactive oxygen/nitrogen species (ROS/RNS) and ADP-ribose (ADPR) is linked to cell death. While these channels are broadly expressed in the CNS, the presence of TRPM2 in neurons remains controversial and more specifically, whether they are expressed in neurons of the hippocampus is an open question. With this in mind, we examined whether functional TRPM2 channels are expressed in this neuronal population. Using a combination of molecular and biochemical approaches, we demonstrated the expression of TRPM2 transcripts and proteins in hippocampal pyramidal neurons. Whole-cell voltage-clamp recordings were subsequently carried out to assess the presence of TRPM2-mediated currents. Application of hydrogen peroxide or peroxynitrite to cultured hippocampal pyramidal neurons activated an inward current that was abolished upon removal of extracellular Ca(2+), a hallmark of TRPM2 activation. When ADPR (300 microM) was included in the patch pipette, a large inward current developed but only when depolarizing voltage ramps were continuously (1/10 s) applied to the membrane. This current exhibited a linear current-voltage relationship and was sensitive to block by TRPM2 antagonists (i.e. clotrimazole, flufenamic acid and N-(p-amylcinnamoyl)anthranilic acid (ACA)). The inductive effect of voltage ramps on the ADPR-dependent current required voltage-dependent Ca(2+) channels (VDCCs) and a rise in [Ca(2+)](i). Consistent with the need for a rise in [Ca(2+)](i), activation of NMDA receptors (NMDARs), which are highly permeable to Ca(2+), was also permissive for current development. Importantly, given the prominent vulnerability of CA1 neurons to free-radical-induced cell death, we confirmed that, with ADPR in the pipette, a brief application of NMDA could evoke a large inward current in CA1 pyramidal neurons from hippocampal slices that was abolished by the removal of extracellular Ca(2+), consistent with TRPM2 activation. Such a current was absent in interneurons of CA1 stratum radiatum. Finally, infection of cultured hippocampal neurons with a TRPM2-specific short hairpin RNA (shRNA(TRPM2)) significantly reduced both the expression of TRPM2 and the amplitude of the ADPR-dependent current. Taken together, these results indicate that hippocampal pyramidal neurons possess functional TRPM2 channels whose activation by ADPR is functionally coupled to VDCCs and NMDARs through a rise in [Ca(2+)](i).

Figure 1. Expression and localization of TRPM2 in mouse brain

A, immunochemical in situ hybridization for TRPM2 in sagittal, adult brain slice. Expanded region illustrates presence of hybridization in hippocampal cell layers. B, RT-PCR of TRPM2 using distinct primer sets (1 and 2) amplifying non-overlapping regions. C, Western blot analysis of TRPM2 expression in crude membrane fraction. D, subcellular localization of TRPM2 by immunofluorescence staining in cultured hippocampal neurons: D₁, anti-TRPM2 (red); D₂, anti-PSD-95 (green); D₃, overlay of D₁ and D₂ including visualization of Hoechst nuclear stain (blue).

Figure 2. Reactive species activate a TRPM2-like current in cultured hippocampal neurons

A, representative whole-cell recording demonstrating activation of a TRPM2-like current upon extracellular H₂O₂ (1 mm) application and block by CLT (10 μm). Bar indicates period of H₂O₂ or CLT application. B, inward current recorded from a single neuron in response to extracellular ONOO⁻ (0.1 mm) application. C, summary bar graph displaying the mean normalized amplitude of the H₂O₂ and ONOO⁻ current remaining after removal of extracellular Ca²⁺ (H₂O₂, n= 8; ONOO⁻, n= 6 or 7) or addition of clotrimazole (CLT) (10 μm) to the ECF (H₂O₂, n= 9).

Figure 3. ADP-ribose (ADPR)-dependent currents in cultured hippocampal neurons

A, averaged data illustrates the time course for the development of inward currents in recordings with ADPR added to the patch pipette (n= 12). In contrast, the holding current necessary to maintain neurons at −60 mV was unchanged in recordings without added ADPR (control, no ADPR, n= 11). B, holding current at start of recordings and after 10 min is shown for each individual recording from each of the recording groups. C, I–V curves from a recording in which ADPR was added to the patch pipette. Traces were taken from voltage-ramps applied at the start of recording (trace 1) and after 10 min (trace 2). The specific contribution from the ADPR-dependent current was derived by subtracting the I–V curve generated at the start from that which was recorded at 10 min (trace 2–1).

Figure 4. Activation of the ADPR-dependent current in hippocampal slices in CA1 pyramidal neurons but not in stratum radiatum interneurons

A, representative whole-cell recording from a CA1 pyramidal neuron demonstrating activation of an ADPR- dependent current (1 mm ADPR in the patch pipette). The amplitude of the current was reduced by extracellular application of 10 μm clotrimazole (CLT) as shown. Ramps were blanked for clarity. B, I–V curves from the same cell. Traces were taken from voltage-ramps applied at the start of recording (trace 1), after 30 min (trace 2) and following perfusion of the slice for 20 min with ACSF containing 10 μm CLT (trace 3). C, summary bar graph demonstrating inhibition of ADPR-dependent current by 10 μm CLT (n= 4). D, summary bar graph of ADPR-dependent currents induced by application of NMDA and abolished by removal of extracellular Ca²⁺ in CA1 pyramidal cells (n= 6 and 8) and stratum radiatum interneurons (n= 7 and 6).

Figure 5. Applied voltage-ramps are necessary for the induction of ADPR-dependent currents

A, average whole-cell currents recorded from neurons with or without added ADPR and in which voltage-ramps were applied continuously or delayed for 10 min. B, summary bar graph illustrating the effects of the various indicated treatments on the amplitude of the ADPR-dependent current measured 10 min after the start of recordings for ramp-induced and 16 min for NMDA-induced. No ADPR (n= 5); ramps: ADPR (n= 13), ADPR + Ni²⁺/Cd²⁺ (n= 6), ADPR + BAPTA (n= 10); NMDA: ADPR (n= 16).

Figure 6. The pharmacological profile of the ADPR-dependent current in cultured hippocampal neurons matches that of TRPM2

Summary data (A) and representative voltage-ramps (B–F) of effect of removing extracellular Ca²⁺ (A and B) or application of either clotrimazole (10 μm) (A and C), flufenamic acid (100 μm) (A and D), ACA 20 μm (A and E) or La³⁺ (100 μm) (A and F) on the maximal ADPR-dependent current.

Figure 7. shRNA knockdown of TRPM2 prevents activation of the ADPR-dependent current

A and B, confocal images from cultures treated as indicated. High magnification (×63 objective) image of the TRPM2 immunosignal (red) (A and B, top left) and GFP fluorescence (green) (A and B, lower right). Anti-NeuN staining (blue) of GFP-positive cells (A and B lower left). Scale bar = 20 μm. C, comparison of fluorescence intensities from cell bodies of WT (uninfected) and neurons infected with shRNA_control and shRNA_TRPM2 (n= 8/group). The TRPM2 fluorescence intensity from cell bodies of the shRNA_TRPM2-treated group is significantly lower than that of the shRNA_control-treated group (also from that of WT cells (GFP negative) from both treatment groups) as indicated by * (difference, P < 0.001, ANOVA; P < 0.001, multiple comparison test (Fisher's least significant difference test)). D, as a control, the NeuN fluorescence intensity was also quantified and used as reference in all 4 groups for the comparison (n= 8/group). No difference in the NeuN expression of these neurons was observed. E, comparison of the ADPR-dependent current induced by NMDA application in neurons treated with shRNA_TRPM2 (n= 9) or shRNA_control (n= 7). The amplitude of the current (normalized to cell capacitance) was determined after 10 min of recordings. The ADPR-dependent current recorded from shRNA_TRPM2-treated neurons was significantly smaller than in shRNA_control-treated neurons (**P < 0.05, Student's t test).