Gain-of-Function Variant TRPC6 A404V Is Associated With Doxorubicin-Related Cardiomyopathy

Abstract

BACKGROUND:

TRPC6 (transient receptor potential canonical 6) channels, encoded by the TRPC6 gene, are widely expressed in cardiomyocytes and play a critical role in maintaining intracellular Ca²⁺ homeostasis. Variants in TRPC6 are associated with chemotherapy-related cardiomyopathy. Specifically, the TRPC6 A404V polymorphism, with a minor (404 V) allele frequency of 12% in the general population, has been identified in patients undergoing anthracycline therapy. However, the underlying mechanisms remain largely unexplored.

METHODS:

Using patch-clamp recordings, Ca²⁺ imaging, computational analysis, and molecular biology techniques, we assessed the effects of doxorubicin and its metabolite, doxorubicinol, on regulating TRPC6 alanine (A) at position 404 replaced by valine (V; A404V) channel expression and function in a heterologous expression system and native cardiac cells.

RESULTS:

Both additive and recessive models demonstrated a significant association between the TRPC6 A404V variant and doxorubicin-related cardiomyopathy. The TRPC6 A404V channel exhibited higher membrane expression levels compared with the wild type (WT) control. Patch-clamp recordings showed that both TRPC6 WT and A404V channels remained mostly inactive at baseline. Application of 50 μmol/L 1-oleoyl acetyl-sn-glycerol (OAG), a TRPC6 activator, significantly increased the inward- and outward-current densities of WT and A404V channels. Furthermore, a 24-hour treatment with 0.5 μmol/L doxorubicin enhanced TRPC6 mRNA expression and potentiated the OAG effects on both WT and A404V channels, with a more pronounced response in A404V channels. Treatment with 0.5 μmol/L doxorubicinol had no effect on OAG-induced current densities in either WT or A404V channels. Doxorubicin effects on intracellular Ca²⁺ levels were confirmed by Ca²⁺ imaging in native cardiac cells. Computational modeling revealed that the A404V mutation induces a conformational change in the OAG-binding pocket, enhancing its interaction with OAG in the A404V protein compared with the WT control.

CONCLUSIONS:

The TRPC6 A404V is a gain-of-function variant that exhibits enhanced activity in the presence of doxorubicin. Therefore, the TRPC6 A404V variant represents a risk factor for anthracycline-induced cardiotoxicity in patients with cancer.

Doxorubicin is a highly effective anthracycline chemotherapeutic agent. Despite its many side effects, doxorubicin is widely used in the treatment of various cancers, including breast cancer, bladder cancer, lymphoma, and leukemia. A serious side effect is cardiotoxicity, resulting in 50% fatality once heart failure is developed.¹ Doxorubicinol is the main metabolite of doxorubicin, which accumulates in the heart and may contribute to cardiotoxicity during doxorubicin therapy.^2,3 The rate of doxorubicin-induced cardiomyopathy is dependent on its cumulative dose, but the threshold doses that would produce doxorubicin-induced cardiomyopathy and heart failure are highly variable, suggesting that genetic factors may contribute to the adverse effects of anti-cancer therapy. However, the precise mechanisms underlying the genetic contribution to anthracycline therapy-induced cardiotoxicity remain largely unknown.

The TRPC6 (transient receptor potential canonical 6) channels, encoded by the TRPC6 gene, are widely expressed in the cardiovascular system, and their expression is controlled by the Ca²⁺-dependent CaN (calcineurin)/NFAT (nuclear factor of activated T-lymphocytes) pathway.⁴ Under normal physiological conditions, the cardiac TRPC6 channel remains largely inactive.^5,6 In pathological conditions, such as pressure overload and neurohormonal stimulation, activation of TRPC6 channels results in membrane depolarization and Ca²⁺ influx, contributing to maladaptive cardiac remodeling and heart failure.^5,6 Ischemia-reperfusion injury induces upregulation of TRPC6, promoting increased Ca²⁺ influx and phosphorylation-mediated activation of Ca²⁺/CaMKII (calmodulin-dependent protein kinase II), thereby aggravating oxidative stress.⁷ In in-vivo models of pressure overload, pharmacological inhibition of TRPC6 ameliorates cardiac fibrosis and contractility. Similarly, in in vivo models of cardiac ischemic-reperfusion injury⁷ and work from our group in a doxorubicin-induced cardiomyopathy mouse model,⁸ genetic deletion of TRPC6 also ameliorated cardiac fibrosis and contractile dysfunction, making TRPC6 a potential therapeutic target for cardiovascular disease.

Because there is significant interindividual variation in doxorubicin-induced cardiomyopathy and heart failure, our group has investigated genetic variation in TRPC6. We have reported that a gain-of-function (GOF) TRPC6 N338S rare mutation was causally associated with congestive heart failure after receiving anthracycline chemotherapy.⁹ TRPC6 A404V variant is also linked to cardiomyopathy and heart failure in patients with and without anthracycline therapy.^10,11 However, the ionic mechanisms and other genetic factors by which the TRPC6 A404V variant induces human cardiomyopathy leading to heart failure remain unclear. Moreover, the genetics of interindividual variation in response to doxorubicin are likely complex. Multiple genes and variants of different effect sizes are reported in the literature with little known of the interactions or overlapping pathways and mechanisms involved in cardiotoxicity.¹² For example, the CBR3 V244M polymorphism (rs1056892, with minor allele frequency about 35% in the general population) in biochemical assays, significantly altered the rate of doxorubicin metabolism to doxorubicinol,¹³ and the higher metabolizing variant has also been associated with anthracycline-related cardiomyopathy.^14–18 Therefore, in this study, we further explored the genetic associations of TRPC6 A404V in our initial genome-wide association study,¹⁶ and examined the activity of TRPC6 A404V in response to 1-oleoyl acetyl-sn-glycerol (OAG), a synthetic diacylglycerol analog, under conditions with and without doxorubicin and doxorubicinol. Our results show that TRPC6 A404V is a GOF variant. The presence of doxorubicin, but not doxorubicinol, further amplifies the OAG effects on TRPC6 alanine (A) at position 404 replaced by valine (V; A404V) channel activation. Hence, patients with cancer carrying the TRPC6 A404V variant may be more susceptible to developing cardiomyopathy, and leading to heart failure following anthracycline chemotherapy.

Methods

This study was approved by the Mayo Clinic institutional review board (no. 15-002789), and informed consent was obtained from all participants. Details of research designs, major resources, and statistical methods used for each experiment are available in the Supplemental Materials.^9,16,19–22 Supporting data are available from the corresponding author on reasonable request.

Results

TRPC6 A404V Is More Strongly Associated With Chemotherapy-Related Cardiomyopathy Under a Recessive Model

In the primary genome-wide association study additive model with quantitative analysis of the maximum decline in left ventricular ejection fraction, the TRPC6 minor allele (404V) was associated with a larger decline in left ventricular ejection fraction compared with the 404A allele, P=0.03, β=−1.07. Next, we explored the association of A404V with the binary outcome of cardiomyopathy, according to the diagnostic criteria,¹⁹ under additive and recessive models. Using an additive model, we observed a significant excess of the 404V allele in cardiomyopathy cases (n=173), 15.9% versus 10.8% in patients that did not experience cardiomyopathy (n=627), P=0.016, and odds ratio, 1.52. Given that our functional validation model of GOF used overexpression of TRPC6, which is more biologically representative of homozygous 404V genotype, we repeated the genetic analysis with a recessive model, comparing the number of homozygous 404V genotypes in cardiomyopathy versus no cardiomyopathy patients. Under the recessive model, we observed 9 homozygous genotypes in 173 cardiomyopathy cases (5.2%) versus 5 homozygous genotypes in 627 patients who did not experience cardiomyopathy (0.8%), P=0.002, odds ratio, 5.95.

Effects of OAG on TRPC6 WT and A404V Channel Currents

We evaluated the effects of 50 μmol/L OAG, a commonly used concentration for TRPC6 channel activation,²³ on whole-cell TRPC6 current densities using a voltage ramp protocol from −100 mV to +100 mV with a holding potential of −60 mV, 48 hours after transfection with TRPC6 wild type (WT) or A404V cDNAs in human embryonic kidney 293 cell (HEK293) cells (Figure 1A and 1D). The inward-current densities at −100 mV and the outward-current densities at +100 mV for TRPC6 WT and A404V channels were comparable at baseline (−4.56±0.98 picoamperes per picofarad [pA/pF] in TRPC6 WT versus −3.16±0.77 pA/pF in A404V at −100 mV, 9.22±1.59 pA/pF in TRPC6 WT versus 11.56±1.77 pA/pF in A404V at+100 mV, n=20 cells; P=0.221). Application of 50 μmol/L OAG significantly enhanced both inward and outward current densities in TRPC6 WT cells, increasing them to −11.65±1.42 pA/pF at −100 mV and 41.52±5.08 pA/pF at +100 mV, respectively, representing a 2.6-fold increase in the inward current densities and a 4.5-fold increase in the outward current densities. The effects of OAG on the A404V channels were about 2.5-fold greater than those on WT (−26.52±3.47 pA/pF at −100 mV and 103.53±14.27 pA/pF at +100 mV, respectively, n=12 cells; P<0.001 versus WT with OAG). Superfusion with 1 μmol/L BI-749327 completely blocked the OAG-induced effect, suggesting that the increase in current was mediated through TRPC6 channels (Figure 1B and 1E). Group results are shown in bar graphs with individual data points (Figure 1C and 1F).

Figure 1.

Effects of 1-oleoyl acetyl-sn-glycerol (OAG) on TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) mutant channel activation. A and D, Whole-cell currents of TRPC6 were elicited from human embryonic kidney 293 cells (HEK293) 48 h after transfection with TRPC6 WT and TRPC6 A404V cDNAs, using a voltage-ramp protocol from −100 mV to +100 mV over 100 ms at a holding potential of −60 mV. At baseline, the A404V mutant has similar inward- and outward-current densities, compared with WT. Application of OAG (50 μmol/L) robustly increased the inward- and outward-currents of TRPC6 WT (A), and the OAG effects were significantly greater in the A404V channels (D). B and E, The time course of OAG effects on the inward-current densities (measured at −100 mV) and the outward-current densities (measured at +100 mV) of TRPC6 WT (B) and A404V channels (E). The effects of OAG on TRPC6 channels were abolished by superfusion with BI-749327 (1 μmol/L), a highly selective TRPC6 inhibitor. C and F, Group results with individual data points and statistical significance are shown in the bar graphs. A paired t test was performed to compare TRPC6 WT and TRPC6 A404V groups before and after OAG treatment, while the Mann-Whitney U test was used to compare TRPC6 WT and TRPC6 A404V groups at baseline and following OAG treatment at both −100 and +100 mV.

Effects of Doxorubicin and Doxorubicinol on OAG-Induced TRPC6 WT and A404V Channel Activation

We further compared the effects of OAG on TRPC6 WT and A404V channel activation after treatment with doxorubicin or its metabolite, doxorubicinol. After a 24-hour transfection with TRPC6 WT or A404V cDNAs followed by an additional 24-hour incubation with 0.5 μmol/L doxorubicin or doxorubicinol in HEK293 cells, the current densities of TRPC6 WT or A404V channels were measured before (at baseline) and after exposure to 50 μmol/L OAG. OAG robustly augmented TRPC6 WT current densities (Figure 2A through 2C), and the OAG effect was significantly greater in A404V channels after incubation with doxorubicin (Figure 2D through 2F), compared with cells without doxorubicin treatment. For instance, doxorubicin potentiated the OAG effects on the inward-current densities at −100 mV and the outward-current densities at +100 mV by 7.3-fold and 9.7-fold, respectively, in TRPC6 WT channels (from −3.48±0.56 pA/pF at −100 mV and 8.77±0.79 pA/pF at +100 mV of baseline to −25.37±3.38 pA/pF at −100 mV and 84.89±11.12 pA/pF at +100 mV with OAG, respectively, n=24 cells; P<0.05). The effects of OAG on TRPC6 A404V channels were significantly augmented by >19.4-fold after treatment of doxorubicin, from −2.27±0.47 pA/pF at −100 mV and 8.68±2.21 pA/pF at +100 mV at baseline to −44.01±5.87 pA/pF at −100 mV and 169.52±25.85 pA/pF at +100 mV with OAG (n=13 cells; P<0.05, A404V with doxorubicin versus A404V at baseline and WT with doxorubicin). Group results are shown in bar graphs with individual data points (Figure 2C and 2F).

Figure 2.

Effects of doxorubicin (DOX) on 1-oleoyl acetyl-sn-glycerol (OAG)-mediated TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) mutant channel activation. A and D, Whole-cell currents of TRPC6 were elicited from human embryonic kidney 293 cells (HEK293) 24 h after transfection with TRPC6 WT and TRPC6 A404V cDNAs, followed by an additional incubation with 0.5 μmol/L DOX for 24 h, using a voltage-ramp protocol from −100 mV to +100 mV over 100 ms at a holding potential of −60 mV. Treatment with DOX did not enhance the OAG (50 µmol/L) effects on the inward-current increase but further potentiated the OAG (50 µmol/L) effects on the outward-current increase in TRPC6 A404V channels, compared with the effects of OAG on TRPC6 WT channels (A). B and E, The time course of 50 µmol/L OAG effects on the inward-current densities (measured at −100 mV) and the outward-current densities (measured at +100 mV) of TRPC6 WT (B) and A404V channels (E) after incubation with DOX. The effects of OAG on TRPC6 channels were abolished by superfusion with BI-749327 (1 μmol/L). C and F, Group results with individual data points and statistical significance are shown in the bar graphs. A paired t test was conducted to evaluate changes in TRPC6 WT and TRPC6 A404V groups before and after OAG treatment, while the Mann-Whitney U test was used to compare groups at baseline and after OAG treatment in the presence of DOX at both −100 and +100 mV.

Treatment with 0.5 μmol/L doxorubicinol showed no significant effects on both inward- and outward current densities of WT and A404V channels at baseline. However, after a 24-hour treatment with doxorubicinol, 50 μmol/L OAG enhanced the inward- and outward-current densities of TRPC6 WT channels by 2.6-fold and 4.7-fold, respectively, from −4.96±1.34 pA/pF to −12.77±1.68 pA/pF at −100 mV and from 8.17±1.73 pA/pF to 38.49±7.42 pA/pF at +100 mV (n=12 cells; P<0.05 versus baseline; Figure 3A through 3C). OAG potentiated A404V channels after treatment with doxorubicinol, by 12.3-fold in the inward current densities and by 13.3-fold in the outward current densities, from −2.26±0.43 pA/pF to −27.77±5.21 pA/pF at −100 mV and from 8.47±0.91 pA/pF to 112.92±16.79 pA/pF at +100 mV (n=10 cells; P<0.05 versus A404V at baseline and WT with OAG; Figure 3D through 3F). Our results indicate that the effects of doxorubicin on OAG-mediated TRPC6 WT and A404V channel activation are much more potent than those of its metabolite, doxorubicinol. Unlike doxorubicin treatment, doxorubicinol treatment did not further potentiate the effects of OAG on TRPC6 WT and A404V channel activation. The details of doxorubicin and doxorubicinol effects on TRPC6 WT and A404V channel activation are summarized in Figure 4.

Figure 3.

Effects of doxorubicinol (DOXol) on 1-oleoyl acetyl-sn-glycerol (OAG)-mediated TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) mutant channel activation. A and D, Whole-cell currents of TRPC6 were recorded from HEK293 cells 24 h after transfection with TRPC6 WT and TRPC6 A404V cDNA, followed by an additional incubation with 0.5 μmol/L DOXol for 24 h, using a voltage-ramp protocol from −100 mV to +100 mV over 100 ms at a holding potential of −60 mV. Treatment with DOXol did not alter the inward- and outward-currents of TRPC6 WT and A404V channels at baseline (before application of 50 µmol/L OAG), but it increased the inward- and outward-currents of TRPC6 WT channels (A) and further potentiated the effects of OAG on the inward- and outward-current increases in TRPC6 A404V channels (D). The effects of OAG on TRPC6 channels were abolished by superfusion with BI-749327 (1 μmol/L). B and E, The time course of 50 µmol/L OAG effects on the inward-current densities (measured at −100 mV) and the outward-current densities (measured at +100 mV) of TRPC6 WT (B) and A404V channels (E). C and F, Group results with individual data points and statistical significance are shown in the bar graphs. A paired t test was performed to compare TRPC6 WT and TRPC6 A404V groups before and after exposure to OAG in cells treated with DOXol. A Student t test was used to compare TRPC6 WT and TRPC6 A404V groups following OAG exposure in cells 24 h after DOXol treatment at −100 mV, as well as to compare TRPC6 WT and TRPC6 A404V baseline currents 24 h after DOXol treatment at +100 mV. The Mann-Whitney U test was applied to compare baseline currents between TRPC6 WT and TRPC6 A404V groups in the presence of DOXol, as well as OAG-treated groups between TRPC6 WT and TRPC6 A404V at +100 mV.

Figure 4.

Summary of 1-oleoyl acetyl-sn-glycerol (OAG) effects on the inward and outward current densities of TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) mutant channels. Bar graphs illustrate the group results of OAG effects on TRPC6 channel inward-current densities measured at −100 mV (upper) and outward-current densities measured at +100 mV (lower) in cells 24 h after incubation with doxorubicin (DOX; 0.5 µmol/L) or doxorubicinol (DOXol; 0.5 µmol/L). Pretreatment with DOX significantly augments the effects of OAG on the inward- and outward-current increases in TRPC6 A404V channels, compared with WT controls, while pretreatment with DOXol fails to further enhance the OAG effects on TRPC6 WT and A404V channels. Neither DOX nor DOXol pretreatment alters the inward- and outward-currents of TRPC6 WT and A404V channels at baseline. The Kruskal-Wallis H test followed by the Dunn posthoc test was applied for comparisons among all data groups, except for those involving TRPC6 A404V at baseline and after exposure to OAG in cells treated with DOX or DOXol, which were analyzed using 1-way ANOVA followed by the Least Significant Difference posthoc test.

Effect of Doxorubicin on Intracellular Ca²⁺ Levels in AC16 Cells Expressing TRPC6 WT and A404V Channels

We measured [Ca²⁺]_i in AC16 cardiac cells 48 hours after cotransfection with TRPC6 WT/GFP (green fluorescent protein) or A404V/GFP cDNAs, at baseline and after exposure to 50 µmol/L OAG. In cells expressing TRPC6 WT, the peak F/F₀ value increased from 1.61±0.04 at baseline to 2.54±0.26 with OAG (P<0.001, n=18). In cells expressing A404V, the peak F/F₀ at baseline was 1.55±0.04 (n=17 cells; P=0.324 versus WT) and 2.14±0.08 with OAG (n=17 cells, P<0.001 versus baseline and WT with OAG; Figure 5, upper panel).

Figure 5.

Effects of 1-oleoyl acetyl-sn-glycerol (OAG) on intracellular Ca²⁺ levels in AC16 cardiac cells expressing TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) mutant channels, with and without pretreatment with doxorubicin (DOX). Upper, TRPC6 WT/GFP (green fluorescent protein) or A404V/GFP cDNAs were transfected with GFP cDNA into AC16 cardiac cells. A 48-h posttransfection, superfusion with OAG (50 µmol/L) significantly increased the peak intracellular Ca²⁺ levels, [Ca²⁺]_i, in cells expressing TRPC6 WT/GFP, with a markedly greater effect observed in cells expressing TRPC6 A404V/GFP (P<0.001 vs WT, the Student t test). Lower: A 24-h pretreatment with DOX (0.5 µmol/L) significantly enhanced the OAG-induced peak [Ca²⁺]_i in cells expressing TRPC6 A404V/GFP compared with TRPC6 WT/GFP (P<0.001 vs A404V, the Student t test). This enhancement was abolished by cotreatment with BI-749327 (0.1 µmol/L).

The effects of doxorubicin on OAG-mediated Ca²⁺ entry were assessed in AC16 cardiac cells 24 hours after TRPC6 WT/GFP or A404V/GFP expression, followed by adding 0.5 μmol/L doxorubicin into the cell culture medium for an additional 24-hour incubation. Adding the same volume of DMSO served as a control. Treatment with doxorubicin enhanced intracellular Ca2+ concentration ([Ca²⁺]_i) levels (peak F/F₀: 2.76±0.30, P<0.001 versus baseline, n=13 cells) in AC16 cells expressing TRPC6 WT/GFP after application of OAG. The effects of doxorubicin on [Ca²⁺]_i levels were significantly greater in AC16 cells expressing A404V/GFP in response to OAG (peak F/F₀: 3.21±0.04, n=16 cells, P<0.001 versus baseline and WT with OAG). There was no statistically significant difference in baseline [Ca²⁺]_i levels regardless of doxorubicin treatment (Figure 5, lower panel). In addition, the effects of OAG were abolished by a 30-minute preincubation with 0.1 μmol/L BI-749327, indicating that the enhanced [Ca²⁺]_i in AC16 cells by OAG was mediated through TRPC6 channel activation.

Effects of Doxorubicin on TRPC6 WT and A404V mRNA and Protein Expression

The mRNA and protein levels of TRPC6 WT and A404V channels were measured in HEK293 cells 24 hours after transfection, with or without an additional 24-hour incubation with 0.5 μmol/L doxorubicin. As shown in Figure 6, at baseline, mRNA levels were comparable between TRPC6 WT and A404V channels, whereas membrane expression of A404V channel was 1.44±0.08-fold higher than that of WT (n=6 replicates, P=0.001), without an associated change in cytoplasmic protein levels. Following doxorubicin treatment, mRNA levels were upregulated by 3.2-fold in the WT channel and 5.5-fold in the A404V channel, while their membrane expressions were increased by 2.4-fold and 3.4-fold, respectively. Cytoplasmic protein levels were no different. Notably, the doxorubicin-induced increases in both mRNA and membrane expression were ≈2-fold greater on the A404V channel than on WT. The pronounced elevation in TRPC6 A404V mRNA expression observed following doxorubicin treatment was attributed to a 2.6-fold increase in NFATc3 nuclear translocation (0.63±0.13 in control versus 1.66±0.31 with doxorubicin treatment; n=6 replicates; P=0.012).

Figure 6.

Regulation of TRPC6 (canonical transient receptor potential 6) wild type (WT) and A404V mRNA and protein expression by doxorubicin (DOX). A, mRNA expression levels are comparable between TRPC6 A404V and WT in HEK293 cells 48 h after transfection. Treatment with DOX results in a 3.2-fold and 5.5-fold increase in the mRNA levels of TRPC6 WT and A404V, respectively (n=6 biological replicates; 1-way ANOVA followed by Holm posthoc test). mRNA expression is represented as relative cDNA amplification (2^–ΔΔCt) normalized to the GAPDH gene. B, Nuclear proteins were extracted from cells 24 h after transfection with TRPC6 WT or A404V cDNA, followed by incubation with 0.5 μmol/L DOX for an additional 24 hours. DOX treatment induced a 2.6-fold increase in NFATc3 nuclear translocation (n=6 biological replicates; Student t test). C, Surface-biotinylated and cytoplasmic TRPC6 proteins (WT and A404V) were detected in HEK293 cells 48 h after transfection. Compared with TRPC6 WT, the A404V shows a significant increase in both membrane expression and the ratio of membrane and cytoplasmic expression (n=6 biological replicates, the Mann-Whitney U test). D, DOX treatment further enhanced those profiles of TRPC6 A404V expression in HEK293 cells (n=6 biological replicates, the Mann-Whitney U test). GAPDH protein served as a cytoplasmic protein marker and loading control.

Molecular Docking Analysis of OAG and TRPC6 Interaction

To analyze the alterations in the OAG-TRPC6 interaction caused by the TRPC6 A404V mutation, we performed molecular docking of OAG on TRPC6 WT and A404V structures. As we have previously reported, OAG binds to a pocket formed by the pore helix, S5, and S6 domains within the same subunit (Figure 7A). Several residues are involved in this interaction, including M638, F641, I642, I682, N712, and N715.⁹ Notably, the A404V mutation induces a conformational change in the OAG-binding pocket, causing a clockwise rotation at M638, M715, and L719 residues by 0.8º, 1.6º, and 1.7º, respectively (Figure 7B). In TRPC6 WT, the I682 residue in the pore helix, along with N712 and M715 in S6, forms conventional and carbon-hydrogen bonds with the oleic acid and acetyl groups of OAG, whereas residues F635, M638, F641, and I642 interact with the glycerol backbone of OAG through multiple hydrophobic interactions (Figure 8A).

Figure 7.

Conformational changes in the 1-oleoyl acetyl-sn-glycerol (OAG)-binding pocket of TRPC6 (canonical transient receptor potential 6) alanine (A) at position 404 replaced by valine (V; A404V) structure. A, The front view of a human TRPC6 structure is generated by PyMOL software based on the cryo-electron microscopy TRPC6 structure (PDB ID no. 6UZ8). The functional TRPC6 channel is composed of 4 homologous subunits. Each subunit contains an intracellular N-terminus, a pre-S1 elbow (blue color), TMD (6 transmembrane domains, green), namely S1–S6, in which S1-S4 constitute the voltage sensor domain and S5-S6 form the ion-conducting pore domain with an intervening pore helix, a highly conserved TRPH (transient receptor potential helix, brown) followed by a TRP reentrant segment (red), and an intracellular C-terminus. In the N-terminus, there are ARs (4 ankyrin repeats, magenta) followed by 9 linker helices (LHs, cyan). In the C-terminus, it contains a vertical helix (VH, yellow) and a horizontal helix (HH, yellow). The A404V mutation is located at the pre-S1 elbow. The pre-S1 elbow and the TRP reentrant segments are halfway embedded in the membrane, playing a pivotal role in channel architecture and function. OAG binds to a pocket (highlighted by a red circle) formed by the pore helix, along with the S5 and S6 domains of the same subunit. B, The conformational alterations in the OAG binding site were analyzed using BIOVIA Discovery Studio Visualizer. TRPC6 WT and TRPC6 A404V mutation have the same OAG-binding site, which is composed of M636, F641, I642, I682, N712, M715, V716, and L719 residues in the pore-helix, S5, and S6 domains within 1 subunit. The A404V mutation causes conformational changes of the OAG-binding site in TRPC6, leading to a clockwise rotation, particularly at M638, M715, and L719 residues by 0.8º, 1.6º, and 1.7º, respectively.

Figure 8.

Molecular binding details of 1-oleoyl acetyl-sn-glycerol (OAG) on TRPC6 (canonical transient receptor potential 6) wild type (WT) and alanine (A) at position 404 replaced by valine (V; A404V) structures. The structure of OAG consists of a glycerol backbone esterified with an oleic acid molecule at the sn-1 position and an acetyl group at the sn-2 position. OAG binds to a groove formed by the S5 and S6 domains, as well as the pore-helix, in both TRPC6 WT (A) and A404V (B). In TRPC6 WT, I682 residue in the pore-helix forms a hydrogen bond with the oleic acid molecule and interacts hydrophobically with the glycerol backbone of OAG. Two additional carbon-hydrogen bonds are formed between N712 and M715 residues in the S6 region of TRPC6 WT and the oleic acid molecule of OAG. The glycerol backbone of OAG forms multiple alkyl/π-alkyl interactions with F635, M638, F641, and I642 residues in S5 of TRPC6 WT. In contrast, more residues and hydrophobic interactions or noncovalent interactions contribute to OAG binding to A404V. Specifically, the M715 residue forms a sulfur-X interaction with the acetyl group of OAG and 2 alkyl/π-alkyl interactions with the glycerol backbone of OAG. In addition, multiple alkyl/π-alkyl interactions are present between the glycerol backbone of OAG and M638, F641 in S5, A681 and I682 in the pore-helix, as well as V716 and L719 in S6 of A404V. Furthermore, the oleic acid molecule of OAG is surrounded by I642, F645, V646, F675, and F679 in S5 through van der Waals forces. These conformational changes result in a lower binding energy and a smaller apparent Kd in OAG-A404V (binding energy: −5.6 kcal/mol; apparent Kd: 0.08 mmol/L), compared with OAG-TRPC6 WT (binding energy: −4.5 kcal/mol; apparent Kd: 0.51 mmol/L).

In TRPC6 A404V, more residues interact with OAG. Specifically, the glycerol backbone of OAG forms hydrophobic interactions with M638 and F641 in S5, I682 in the pore-helix, and V716 and L719 residues in S6. M715 forms a sulfur-X interaction with the acetyl group of OAG, as well as 2 hydrophobic interactions with the glycerol backbone. Furthermore, the oleic acid molecule of OAG engaged with the I642, F645, V646, F675, and F679 residues in S5 via van der Waals forces (Figure 8B). These structural alterations enhance the interaction between OAG and TRPC6 A404V, as reflected by the decreased binding energy of OAG (−5.6 kcal/mol for TRPC6 A404V versus −4.5 kcal/mol for WT) and the lower apparent dissociation rate constant (K_d) of OAG (0.08 mmol/L for TRPC6 A404V versus 0.51 mmol/L for WT), representing reductions of 0.8-fold and 6.4-fold, respectively, in TRPC6 A404V.

Discussion

In this study, we evaluated the expression and function of the TRPC6 A404V channel in HEK293 cells and AC16 cardiac cells. We have several important findings. First, the TRPC6 A404V channel has a greater response to OAG activation, enhancing the TRPC6 A404V current densities and Ca²⁺ influx in HEK293 cells and AC16 cardiac cells. Second, TRPC6 shows increased mRNA and membrane expression levels, which are further upregulated in response to doxorubicin treatment. Third, the presence of doxorubicin, but not its metabolite doxorubicinol, potentiated the effects of OAG on TRPC6 WT and A404V current densities. Fourth, molecular docking results indicate a higher binding affinity of OAG for TRPC6 A404V channels, confirming that TRPC6 A404V is a GOF variant.

TRPC6 is highly expressed in cardiovascular tissues and plays an important role in cardiovascular function by mediating Ca²⁺ entry into vascular smooth muscle cells and cardiomyocytes, which is crucial for the regulation of vascular tone, cardiac contractility, and cell proliferation.²⁴ Dysregulation of TRPC6 expression or function has been implicated in the processes of cardiovascular diseases, such as hypertension, heart failure, and cardiac hypertrophy.^5,6,25,26 In our previously published genetic studies, the valine substitution at residue 404 (404V) was associated with the greatest decline in cardiac systolic function, both in our initial genome-wide association study of patients with breast cancer treated with doxorubicin and in noncancer patients with heart failure of various causes from the Mayo Clinic Biobank.^11,16 The TRPC6 A404V variant is a common polymorphism. The minor allele (404V) has a frequency of ≈12% in European and South Asian populations, with lower but still notable frequencies of around 2% in African/Black populations and 5% in East Asian populations, according to the GnomAD database.²⁷ In the current study, further analysis showed the strongest association under a recessive model using a binary outcome of cardiomyopathy, which is in close alignment with TRPC6 overexpression used in the functional model. The adequate sample size allowed for the detection of statistically significant effects, supporting the validity of the findings under this model. Therefore, the clinical translational finding of this study is that increased intracellular Ca²⁺, mediated by TRPC6, is a potentially treatable risk factor for cardiomyopathy and heart failure that could impact patients with 404V homozygosity. However, the fact that only some individuals with the 404V variant develop cardiotoxicity suggests incomplete penetrance, likely influenced by additional genetic, environmental, or lifestyle factors.

Doxorubicin is a first-generation anthracycline drug effective against various cancers by inhibiting DNA topoisomerase II. It is primarily converted to doxorubicinol by carbonyl reductase enzymes CBR1 (carbonyl reductase) and CBR3, as well as alcohol dehydrogenase in the liver. It is well-known that doxorubicin produces severe cardiotoxicity. In addition, doxorubicinol has been implicated in the cardiotoxic effects associated with doxorubicin treatment,²⁸ with the CBR3 V244M polymorphism leading to significantly different rates of doxorubicin metabolism.¹³ The CBR3 V244M polymorphism has also been associated with doxorubicin-related heart failure by multiple independent groups,^14,18,29 including our own.^10,15 Based on these studies and the known complexity of doxorubicin-related cardiotoxicity,¹² we compared the effects of doxorubicin and doxorubicinol in the TRPC6 A404V channels. In these experiments, we found that a 24-hour treatment with 0.5 µmol/L doxorubicin enhanced the effects of OAG on TRPC6 WT channel currents, with an even greater increment observed in TRPC6 A404V channels. In contrast, the treatment with doxorubicinol failed to further potentiate the effects of OAG on TRPC6 WT and A404V channel current densities. These results imply that the underlying mechanisms regarding the regulation of TRPC6 channel activities by doxorubicin and doxorubicinol are different in the heterologous expression system and that the genetic associations of CBR3 and TRPC6 with doxorubicin-induced cardiotoxicity may be through independent mechanisms.

Human TRPC6 channel is composed of 4 homologous subunits, each containing 6 transmembrane domains (S1–S6) and cytoplasmic N- and C-termini. A notable structural feature of TRPC6 is the pre-S1 elbow, a 3-helix region located before the first transmembrane domain (S1).³⁰ It has been reported that the pre-S1 elbow is embedded in the lipid bilayer and interacts hydrophobically with the intracellular portion of the S1 in the same subunits, playing an essential role for TRPC6 channel function.³⁰ Moreover, the pre-S1 elbow also interacts with the TRP reentrant loop of the adjacent subunit, which is essential for the structural stability and gating of the channel.³⁰ The A404 residue is located at the pre-S1 elbow. In this study, we found that a single mutation at residue A404V causes a 2.5-fold greater response to OAG compared with WT control; however, their baseline current densities are comparable, confirming that cardiac TRPC6 channels are largely inactive without stimulation.^5,6

The molecular mechanism by which OAG activates TRPC6 channels is revealed by molecular docking analysis, which shows that OAG binds to the pore helix as well as the S5 and S6 domains of TRPC6 through hydrogen bonds and hydrophobic interactions.^9,31 Structural analysis of the TRPC6 protein reveals that the A404V mutation induces a clockwise rotation of the OAG-binding pocket. Since the precise 3-dimensional structure is critical for protein function, it is not surprising that such small alterations in OAG binding structure can influence the interactions between TRPC6 A404V and OAG, leading to reduced binding energy and an increased Kd value for OAG binding. The exact cause of the altered 3-dimensional structure of the OAG binding pocket is unknown. It may result from the A404V substitution, where the replacement of alanine with the bulkier valine residue could modify the interaction of pre-S1 elbow with the S6 domain in the adjacent subunit, subsequently impacting the overall structure of the OAG binding site. A similar result was observed with another GOF TRPC6 rare mutant, N338S (minor allele frequency of 0.00002 in the gnomAD database), identified in a 32-year-old patient with breast cancer who developed heart failure following anthracycline chemotherapy,⁹ although its effect was ≈1-fold lower than the A404V variant. The N338 residue is situated at the start of the link helix 6, near the distal end of the horizontal helix domain, which bends ≈90° toward the channel’s central axis to form a cross-shaped structure aligned parallel to the membrane surface. It has been reported that small conformational shifts in the horizontal helix can thus be transmitted to the gate, modulating channel opening and ion conduction.³² Hence, computational analysis represents a rapid and effective approach for the functional prediction of TRPC6 missense variants, providing a valuable strategy for identifying high-risk variants associated with chemotherapy-induced cardiotoxicity.

An additional observation associated with the GOF effect of the TRPC6 A404V channel is a marked increase in its mRNA expression and surface localization, without corresponding changes in cytoplasmic protein levels compared with WT control. These effects are further amplified following doxorubicin treatment. TRPC6 mRNA transcription is regulated by the CaN/NFAT signaling pathway and reactive oxygen species, both of which are activated by doxorubicin in cardiomyocytes.^33–35 We confirmed that doxorubicin treatment significantly increased NFATc3 nuclear translocation and TRPC6 mRNA expression in cells expressing TRPC6 A404V compared with those expressing TRPC6 WT. The enhanced NFATc3 nuclear translocation observed in cells expressing TRPC6 A404V channels is likely due to increased Ca²⁺ influx through the GOF effect of A404V channels. Despite comparable mRNA expression levels, interestingly, the TRPC6 A404V exhibited greater membrane localization at baseline than the WT control. The mechanisms underlying an increase in TRPC6 A404V surface localization are unclear. One explanation is that valine contains a branched side chain with 3 carbon atoms, which increases its hydrophobicity and potentially enhances TRPC6 A404V stabilization within lipid environments.³⁶ Using a homologous model of TRPC6, we confirmed that the A404V channel has a lower energy of protein minimization, which is thermodynamically favorable for the stabilization of the TRPC6 protein.³⁷

In summary, by combining patch-clamp recordings, intracellular Ca²⁺ imaging, in silico analysis, and molecular biological approaches, we provide compelling evidence that TRPC6 A404V is a GOF variant with enhanced response to OAG stimulation. Doxorubicin, but not doxorubicinol, potentiates the OAG effects on TRPC6 A404V channel activation. However, the physiological relevance remains limited, since functional assessment was performed only in cultured HEK293 and AC16 cells and not in primary cardiomyocytes or induced pluripotent stem cell-derived cardiac cells. Future studies using knock-in animal models to investigate the TRPC6 A404V mutation would provide critical in vivo validation. In addition, it should be emphasized that numerous genes can modify the responses to environmental stressors and lifestyles, and heart failure in these individuals may result from any number of such stressors that trigger or enhance TRPC6 expression or activity. Nevertheless, the TRPC6 404 V homozygosity may represent a risk factor for the development of heart failure in patients both with and without anthracycline chemotherapy.

Article Information

Acknowledgment

This work is dedicated to the memory of Professor Wen-Ping Jiang, whose mentorship, insight, and unwavering support during the early stages of my research career will always be remembered.

Sources of Funding

This work is supported by grants from the National Heart, Lung, and Blood Institute (R01HL169268 and R01HL161821), US Department of Defense (W81XWH22-1-0288/PR210385), and Department of Cardiovascular Medicine (Prospective Research Award 2023), Mayo Clinic, MN.

Disclosures

None.

Supplemental Material

Supplemental Methods