5-HT3 Receptors on Mitochondria Influence Mitochondrial Function

Abstract

The 5-hydroxytryptamine 3 (5-HT₃) receptor belongs to the pentameric ligand-gated cation channel superfamily. Humans have five different 5-HT₃ receptor subunits: A to E. The 5-HT₃ receptors are located on the cell membrane, but a previous study suggested that mitochondria could also contain A subunits. In this article, we explored the distribution of 5-HT₃ receptor subunits in intracellular and cell-free mitochondria. Organelle prediction software supported the localization of the A and E subunits on the inner membrane of the mitochondria. We transiently transfected HEK293T cells that do not natively express the 5-HT₃ receptor with an epitope and fluorescent protein-tagged 5HT3A and 5HT3E subunits. Fluorescence microscopy and cell fractionation indicated that both subunits, A and E, localized to the mitochondria, while transmission electron microscopy revealed the location of the subunits on the mitochondrial inner membrane, where they could form heteromeric complexes. Cell-free mitochondria isolated from cell culture media colocalized with the fluorescent signal for A subunits. The presence of A and E subunits influenced changes in the membrane potential and mitochondrial oxygen consumption rates upon exposure to serotonin; this was inhibited by pre-treatment with ondansetron. Therefore, it is likely that the 5-HT₃ receptors present on mitochondria directly impact mitochondrial function and that this may have therapeutic implications.

Keywords: 5-hydroxytryptamine 3 (5-HT3) receptor; immunogold staining; ion channel subunits; ligand-gated ion channels; mitochondria; ondansetron; serotonin type 3 receptor.

Figure 1

Expression of 5HT3A and 5HT3E subunits in transiently transfected HEK293T cells. (a) Schematic of 5HT3A^{mCherry-c-Myc} construct used for transfections. (b) Fixed HEK293T cells transiently transfected with 5HT3A^{mCherry-c-Myc} (red channel) and stained with TOM22 antibody (secondary antibody-Alexafluor 488, green channel), and merged image in which yellow arrows point to mitochondria overlapping with subunit 5HT3A^{mCherry-c-Myc} signal. Scale bar is 10 μm.

Figure 2

5HT3A subunits are present on mitochondria isolated from transfected HEK293T cells. (a) Mitochondria fractions separated in OptiPrep–sucrose gradient (top to bottom order in overlaid crude mitochondria pellet: 10%, 30%, and 50%) and the immunoblots of the whole cell lysate (W), microsome fraction (M) and mitochondrial fractions (1–4) probed with markers for plasma membrane (PM; Na⁺/K⁺ ATPase), endoplasmic reticulum (ER; disulfide isomerase, PDI), Golgi apparatus (GA; Golgin 97), and mitochondria (Mt; TOM22). Full-length immunoblots are shown in Supplementary Figure S10. (b) Immunoblot of mitochondrial fractions 2 and 3, isolated from cells transfected with 5HT3A^{mCherry-c-Myc} and probed with c-Myc and TOM22 antibodies. Full-length immunoblots are shown in Supplementary Figure S11a,b. (c) Partially dispersed pelleted mitochondrial fraction 3 isolated from 5HT3A^{mCherry-c-Myc}-transfected cells. Red channel shows m-Cherry signal, while green channel shows TOM22 antibody detected with Alexafluor 488 with yellow signal overlap in the merged image. Scale bar is 5 μm.

Figure 3

5-HT₃ receptor subunits present on the inner membrane of mitochondria. Transmission electron micrograph images of mitochondria in 5HT3A^{mCherry-c-Myc} (a,b) or 5HT3E^mCherry-HA (c,d) and 5HT3A^c-Myc+ 5HT3E^mCherry-HA (e,f) transiently transfected HEK293T cells. (a) Negative control of mitochondria probed with secondary mouse 15 nm gold nanoparticles only, and gold particles are pointed to with blue-colored arrow heads. (b) Mitochondria probed with mouse c-Myc antibody to detect 5HT3A^{mCherry-c-Myc} using secondary mouse 15 nm gold nanoparticles. (c) Negative control of mitochondria probed with secondary rabbit 6 nm gold nanoparticles. (d) Mitochondria probed with rabbit mCherry antibody to detect 5HT3E^mCherry-HA using secondary rabbit 6 nm gold nanoparticles. (e) Negative control of mitochondria probed with secondary rabbit 6 nm and secondary mouse 15 nm gold nanoparticles. (f) Mitochondria probed with primary mouse c-Myc antibody to detect 5HT3A^c-Myc and rabbit mCherry antibody to detect 5HT3E^mCherry-HA using secondary mouse 15 nm and rabbit 6 nm gold nanoparticles. Gold nanoparticles that stained 5HT3A^{mCherry-c-Myc}, 5HT3E^mCherry-HA, and 5HT3A^c-Myc + 5HT3E^mCherry-HA subunits are pointed to with green, maroon, and yellow arrowheads, respectively. Scale bars are indicated in each figure.

Figure 4

5HT3A subunit on SH SY5Y mitochondria and cell membrane. (a) Transmission electron micrograph image of SH SY5Y cells (negative staining) probed with anti-mouse antisera (secondary) tagged with 15 nm gold nanoparticles to detect non-specific binding of the antibody. Scale bar 600 nm. (b) Transmission electron micrograph image of SH SY5Y cells probed with mouse 5HT3A antibody (primary) and anti-mouse antisera (secondary) tagged with 15 nm gold nanoparticle to detect native 5HT3A subunits. The 15 nm gold nanoparticles on the mitochondria and plasma membrane are indicated by green and maroon arrowheads, respectively. Scale bar 1 μm. Antibody details can be found in Supplementary Table S2.

Figure 5

Cell-free mitochondria carry 5HT3A subunits. (a) Immunoblots of cell-free mitochondria collected from cell culture media of HEK293T cells transiently transfected with 5HT3A^{mCherry-c-Myc} and probed with c-Myc epitope tag antibody (c-Myc) to detect 5HT3A subunits and TOM22 antibody to detect mitochondrial membranes. Full-length immunoblots can be viewed in Supplementary Figure S11c. (b) Pelleted clumps of cell-free mitochondria collected from the cell culture media of 5HT3A^{mCherry-c-Myc}-transfected HEK293T cells. Red channel shows 5HT3A, and green channel shows TOM22 antibody-stained mitochondria. The merged image shows mitochondria with 5HT3A subunits. Scale bar is 5 μm.

Figure 6

5-HT₃ receptors affect mitochondrial membrane potential. (a) Schematic timeline representation of the TMRM and drug treatment for both HEK293T cells and isolated mitochondria. (b) Mitochondrial membrane potential in HEK293T cells (250 cells/well) mock-transfected or transiently transfected with 5HT3A or 5HT3E or both 5HT3A+5HT3E subunits and treated with a vehicle (Milli-Q water; control), serotonin alone, or ondansetron plus serotonin. The fluorescence values obtained when measuring TMRM fluorescence were corrected for the cell number. (c) Mitochondrial membrane potential in purified mitochondria (500 µg/well) obtained from HEK293T cells mock-transfected or transfected with 5HT3A or 5HT3E or both 5HT3A+5HT3E subunits. Mitochondria were treated with a vehicle (Milli-Q water; control) or serotonin alone or ondansetron plus serotonin. For both b and c, all samples were treated with FCCP after drug treatment as a depolarization control, and data were analyzed via two-way ANOVA, followed by Dunnett’s multiple comparisons test (n = 6, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns—not significant).

Figure 7

5-HT₃ receptor subunits affect cellular oxygen consumption rates. (a) Schematic of the drug treatment of HEK293T cells and fluorescence measurement timings. (b–e) Oxygen consumption rate in HEK293T cells (250 cells per well) mock-transfected or transiently transfected with 5HT3A, 5HT3E, or both 5HT3A and 5HT3E subunits. HEK293T cells were treated with vehicle (Milli-Q water; control) or serotonin (30 μM) alone, ondansetron (3 nM) and serotonin (30 μM), or FCCP alone (10 μM). (b) HEK293T cells mock transfected. (c) HEK293T cells transiently transfected with 5HT3A. (d) HEK293T cells transiently transfected with 5HT3E. (e) HEK293T cells transiently transfected with 5HT3A + 5HT3E. (f) Comparison of oxygen consumption rates of mock-transfected and 5HT3A and/or 5HT3E transfected HEK293T cells at 12 min timepoint. All assays were analyzed by two-way ANOVA, followed with Dunnett’s multiple comparison test (n = 3; ** p < 0.01, **** p < 0.0001, ns—not significant, green lines indicate time course differences between control and serotonin and blue lines between control and ondansetron in (b) to (e)).

Figure 8

5-HT₃ receptor subunits affect mitochondrial respiration. (a) Schematic of the drug treatment of isolated mitochondria and fluorescence measurement timings. (b–e) Oxygen consumption rate in purified mitochondria (500 μg per well) isolated from HEK293T cells mock-transfected or transiently transfected with 5HT3A or 5HT3E or both 5HT3A and 5HT3E subunits. Mitochondria were treated with vehicle (Milli-Q water; control) or serotonin (30 μM) alone or ondansetron (3 nM) plus serotonin (30 μM) or FCCP alone (10 μM). (b) Mitochondria from mock-transfected HEK293T cells. (c) Mitochondria from HEK293T cells transiently transfected with 5HT3A. (d) Mitochondria from HEK293T cells transiently transfected with 5HT3E. (e) Mitochondria from HEK293T cells transiently transfected with 5HT3A + 5HT3E. (f) Comparison of oxygen consumption rates of isolated mitochondria at 12 min timepoint. All assays were analyzed via two-way ANOVA, followed with Dunnett’s multiple comparison test (n = 3; ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns—not significant, black lines in parts c and e highlight the time course significant differences between control and serotonin or ondansetron treatments).