TRPM7-mediated Ca²⁺ signals confer fibrogenesis in human Atrial Fibrillation

Human Cardiac Tissue Sample Collection and fibroblasts isolation

Myocardial samples of the right atria were collected during cardiac surgery. All procedures involving human tissue use were approved by the institutional review boards of the University of Connecticut Health Center. Atrial fibroblasts were isolated by collagenase (see details in the supplementary methods). Freshly isolated fibroblasts from AF and NSR patients were used for patch-clamp experiment to compare the TRPM7 current amplitude.

Molecular Biology

RT-PCR was performed as we previously reported ⁴¹ by using the primers shown in Online Table II. TRPM7-specific shRNAs were designed and constructed into lenti-viral delivery system. The efficiency of the shRNAs was tested in the TRPM7 over-expression HEK-293 cells. The one with higher efficiency was used in the experiments (see the Online Data Supplement)

Electrophysiology

Whole-cell and single-channel currents were recorded as we reported previously ^{42, 43}. Internal Mg²⁺-free solution was used for whole-cell current recording and inside-out configuration was applied for single channel current recording. For details, see the Online Data Supplement.

Ratio Ca²⁺ imaging experiments

Ca²⁺ influx was measured using ratio Ca²⁺ imaging system and analyzed using NIS-Elements (Nikon). The F340/F380 ratio in the presence of 20 mM Ca²⁺ was normalized to that of 1 μM Ionomycin (Iono) elicited Ca²⁺ signal (see details in supplementary methods). Ca²⁺ influx was also measured with/without store depletion by 2 μM thapsigargin (Tg) in some experiments as specified in text.

Immunostaining

Fibroblasts were fixed by 4% formaldehyde and immmunostained with α-SMA antibody. The immunostained cells were analyzed using a Zeiss LSM 510 confocal microscope. Details are described in the Online Data Supplement.

Western Blot

The protein expression levels of TRPM7 and α-SMA were evaluated by Western blot experiments. GAPDH was used as a loading control (the Online Data Supplement).

Data Analysis

Pooled data are presented as mean ± SEM. Dose–response curves were fitted by an equation of the form E=E_max{1/[1+(EC₅₀/C)ⁿ]}, where E is the effect at concentration C, E_max is maximal effect, EC₅₀ is the concentration for half-maximal effect and n is the Hill coefficient. Statistical comparisons were made using two-way ANALYSIS of variance (ANOVA) and two-tailed t-test with Bonferroni correction. Fisher’s exact test was used for analysis of percentage differences for patients’ database. P < 0.05 indicated statistical significance.

TRPM7 is the major Ca²⁺-permeable channel in human atrial fibroblasts

To understand which Ca²⁺ permeable channel is responsible for Ca²⁺ entry in cardiac fibroblasts, we first investigated the expression of Ca²⁺-permeable TRP channels. Among the 27 human TRP channel genes ³³, we found that TRPM7 is abundantly expressed in human atrial fibroblasts (Online Figure I). To study whether TRPM7 constitutes functional channels in the fibroblasts, we applied voltage-clamp to investigate the current characteristics. As shown in Figure 1, a TRPM7-like current was readily elicited by a voltage ramp protocol ranging from -120 to +100 mV. The current-voltage relation (I-V relation) with small inward current and strong outward rectification (Figure 1A) is similar to the typical characteristics of the recombinant TRPM7 ^{36, 37}. Heterologous expressed TRPM7 exhibits several unique features, including single channel conductance, potentiation by low external pH, inhibition by low concentrations of 2-aminoethoxydiphenyl borate (2-APB) but potentiation by high concentration of 2-APB ^{43, 44}. We then tested if the endogenous TRPM7-like current in cardiac fibroblasts has similar properties. The single channel conductance of TRPM7-like current (Figure 1B) is similar to that of TRPM7 over-expressed in CHO cells ⁴³. TRPM7-like current was significantly potentiated by low extracellular pH (Figure 1C) with an EC₅₀ of 3.9 pH unit (Online Figure IIB). Like TRPM7, TRPM7-like current was inhibited by low concentrations of 2-APB with an IC₅₀ of 32.1 μM (Figure 1E-F). High concentrations of 2-APB enhanced TRPM7-like current (Figure 1E), but the concentration-dependence of the effect was not assessed because of poor solubility of 2-APB at high concentrations. All of these features of the endogenous TRPM7-like currents in human atrial fibroblasts are indistinguishable from those of exogenously expressed TRPM7 ^{42, 43}, suggesting that the TRPM7-like current in the human cardiac fibroblasts is encoded by TRPM7.

To further confirm that TRPM7 constitutes TRPM7-like currents, we applied TRPM7-specific shRNA to the fibroblasts (Figure 2). The efficiency of TRPM7-shRNA was first examined in HEK-293 cells over-expressing human TRPM7 (Figure 2A-B). The current amplitude of TRPM7 was decreased by ~70% after TRPM7-shRNA treatment. Similarly, after treatment for 5 days, the current amplitude of TRPM7-like currents in human fibroblasts was significantly reduced by TRPM7-shRNA in comparison with the cells treated with Scr-shRNA (Figure 2C-D), indicating that TRPM7 is indeed the molecule encoding TRPM7-like currents in human atrial fibroblasts. Based on these findings, we refer to the TRPM7-like currents as “TRPM7” in the following text.

TRPM7 is responsible for Ca²⁺entry in human atrial fibroblasts

We next asked whether TRPM7 is responsible for Ca²⁺ entry in atrial fibroblasts. Using ratio Ca²⁺ imaging measurements, we found that Ca²⁺ influx was decreased by 62% after fibroblasts were treated with TRPM7-shRNA for 5 days, indicating that TRPM7 is the major Ca²⁺-permeable channel in human atrial fibroblasts (Figure 2E-F). To elucidate whether there are other Ca²⁺-permeable channels which may contribute to Ca²⁺ influx in human atrial fibroblasts, we used the conditions specific for TRPV2 ⁴⁵, TRPV4 ⁴⁶ or TRPC6 ^{47, 48} current recording, and tested if functional currents of these channels are present in human atrial fibroblasts. As shown in Figure 3, TRPV2-like currents could not be elicited by 200 μM 2-APB in human atrial fibroblasts (Figure 3A), whereas 2-APB readily activated TRPV2 overexpressed in HEK293 cells (Figure 3B-C). The conditions used for activating TRPV4 expressed in HEK293 cells (Figure 3E-F) failed to induce any TRPV4-like current in atrial fibroblasts. TRPC6 can be elicited by activation of G_q-like receptor activation (Figure 3H-I). However, neither using LPA to activate EDG7 receptor (Figure 3G) nor including GTPγS in the pipette solution (Online Figure III) was effective in inducing TRPC6-like current in atrial fibroblasts. These results, summarized in Figure 3, indicate that TRPV2, TRPV4, and TRPC6 do not constitute functional currents even though their expression was detectable by RT-PCR (Online Figure I). Therefore, it is unlikely that TRPV2, TRPV4 or TRPC6 contribute to Ca²⁺ entry in human atrial fibroblasts.

It appears that TRPC1 is abundantly expressed at mRNA level (Online Figure I) in human atrial fibroblasts. Since TRPC1 may form heteromeric channels with TRPC4 or TRPC5 ⁴⁹, we attempted to record heteromeric currents with I-V relations similar to those of TRPC1/TRPC4 (or TRPC1/TRPC5). However, including GTPγs in the pipette solution failed to induce any TRPC1/TRPC4- or TRPC1/TRPC5-like current (data not shown), suggesting that TRPC1 may not be able to form functional heteromeric channels in atrial fibroblasts. Knocking down hTRPC1 by TRPC1 siRNA did not influence Ca²⁺ influx or fibroblast differentiation (Online Figure IV), further indicating that TRPC1 may not be able to contribute Ca²⁺ influx in human atrial fibroblasts.

We next investigated whether store depletion-activated Ca²⁺ channel (I_CRAC) may influence Ca²⁺ entry in atrial fibroblasts. As shown in Figure 5, store-depletion by 5 μM thapsigargin (Tg) did not induce any current, suggesting that store depletion activated channels are not functionally expressed in human atrial fibroblasts. Moreover, Ca²⁺ influx was not changed by Tg, further suggesting that store-depletion does not contribute to Ca²⁺ signals in human atrial fibroblasts. We also investigated whether TRPV2, TRPV4, TRPC1/TRPC4, TRPC6 and I_CRAC currents are present in fibroblasts from AF patients, and found that TRPM7 current is the only current that we were able to record in fibroblasts from both NSR and AF patients. Therefore, it appears that endogenous TRPM7 is the major channel which mediates Ca²⁺ entry in human atrial fibroblasts.

TRPM7 is markedly up-regulated in atrial fibroblasts from AF patients

It has been shown that fibrosis associated structural remodeling is a major contributor to AF ⁷. However, the Ca²⁺ signaling mechanism in fibrogenesis is unknown. Our findings that TRPM7 is the major Ca²⁺-permeable channel in human atrial fibroblasts provided us an excellent opportunity to explore how Ca²⁺ signals are involved in fibrosis associated AF. The left atrial size of AF patients was similar to that in NSR patients among those enrolled in the current study (Online Table I), although a larger group of AF and NSR patients, including those who were not enrolled in the fibroblast study, showed a larger left atrial size in the AF vs. NSR patients (data not shown). We used freshly isolated fibroblasts from biopsy specimens for current recording. This eliminated cell culture related factors which could interfere with the experimental results. We found that TRPM7 currents recorded from freshly isolated fibroblasts in AF patients were 3- to 5-fold larger than those from NSR patients (Figure 4). Both inward and outward currents were increased significantly. Since measurement of outward current amplitude gives a more accurate assessment, we used outward current amplitude measured at +100 mV for comparison (Figure 4C-D). The average TRPM7 current density was 32.5 ± 2.3 pA/pF in fibroblasts from AF patients (88 cells from 7 patients) and 11.1±0.5 pA/pF in fibroblasts from NSR patients (87 cells from 8 patients), respectively. The increase in TRPM7 current density appears to be resulted from increased membrane expression of TRPM7 channel proteins, as the single channel conductance and open probability of TRPM7 were indistinguishable between the NSR fibroblasts and AF fibroblasts (Online Figure VI). We were not able to do western blot experiments to determine TRPM7 protein levels in NSR and AF fibroblasts due to the limited amount of tissues and limited number of fibroblasts. However, we observed that whereas we were able to obtained single channel recordings from more than 95% of patches in the AF fibroblasts, less than 50% of patches from NSR fibroblasts displayed single channel openings, suggesting that membrane protein level of TRPM7 was higher in AF fibroblasts than that in NSR fibroblasts.

To study if the increased TRPM7 currents result in enhanced Ca²⁺ influx, we used ratio Ca²⁺ imaging method to compare the Ca²⁺ entry through TRPM7 in fibroblasts isolated from AF and NSR patients. Since TRPM7 channels conduct small inward currents and are constitutively active, we used 20 mM Ca²⁺ in order to obtain better signal/noise ratio. In addition, cells were briefly exposed to a nominally divalent free solution (DVF) in order to eliminate any blockade of TRPM7 by external Mg²⁺. As shown in Figure 5, the peak Ca²⁺ influx in fibroblasts from AF patients was 10-fold greater than Ca²⁺ influx in the fibroblasts from NSR patients. This result not only further indicates that TRPM7 is the major Ca²⁺-permeable channel responsible for Ca²⁺ influx in human atrial fibroblasts, but also strongly implies that TRPM7-mediated Ca²⁺ signals may play an important role in AF-associated fibrogenesis.

TRPM7 is required for differentiation of human atrial fibroblasts

As the myofibroflast is the major cell type that synthesizes growth factors, cytokines, chemokines, and extracellular matrix (ECM) proteins, and plays a pivotal role in fibrogenesis ^{13, 19}, we tested whether there was a higher percentage of myofibroblasts in the cells isolated from atrial samples from AF patients than those from NSR patients. As shown in Figure 6, after 24 hrs culture, without TGF-β1 treatment, cells from AF patients displayed 44% myofibroblasts, whereas only 16% cells from NSR patients were myofibroblats. Such data suggest that fibroblasts from AF patients are more prone to differentiation into myofibroblasts. The increased vulnerability to differentiation in the fibroblasts from AF patients might result from an up-regulated TRPM7 level. After TGF-β1 (10 ng/ml) induction, 90% of fibroblasts differentiated into myofibroblasts in both NSR and AF groups.

To determine the contribution of TRPM7 to basal differentiation of fibroblasts from AF patients, AF fibroblasts were treated with TRPM7-shRNA or Scr-shRNA in 1% FBS media after atrial fibroblasts attached to the coverslips (by culturing for 12 hrs in 10% FBS media) for 72 hrs. Fibroblasts were then cultured in 10% FBS media for 24 hr, followed by Ca²⁺ influx and differentiation assessment. As shown in Figure 6C-D, TRPM7-shRNA significantly decreased Ca²⁺ influx and inhibited basal AF fibroblast differentiation, indicating that TRPM7-mediated Ca²⁺ entry contributes to basal differentiation of AF fibroblasts.

We further investigated whether TRPM7 is involved in TGF-β1 induced differentiation. Human atrial fibroblasts from NSR patients were treated with TRPM7-shRNA and scramble shRNA (Scr-shRNA). TGF-β1 was then applied to induce differentiation after cells were serum starved for 24 hours. As shown in Figure 7, without TGF-β1 (10 ng/ml) treatment, there were 30.0% myofibroblasts in the Scr-shRNA treated group and 25.4% myofibroblasts in the TRPM7-shRNA treated group. This higher basal differentiation of NSR fibroblasts in comparison with that in Fig. 6 was due to that the fibroblasts were culture in 10% FBS media for more than 84 hrs, and there is a time-dependent increase in basal differentiation rate when fibroblasts were cultured in 10% FBS media (Online Figure VII). After TGF-β1 stimulation, the differentiated myofibroblasts in the Scr-shRNA treated group increased to 61.0%, whereas, knocking down of TRPM7 by TRPM7-shRNA significantly reduced the differentiated myofibroblasts to 34.4%, a level similar to the control group (Figure 7B). The loss of sensitivity to TGF-β1 in TRPM7-shRNA treated fibroblasts suggests that TRPM7 plays a role in TGF-β1 induce atrial fibroblast differentiation.

To understand how TRPM7 may be involved in fibroblast differentiation, we quantified changes of α-SMA protein in scramble- and TRPM7-shRNA treated cells by Western blot. This experiment was performed by using mouse fibroblasts due to the limited number of human fibroblasts. To rule out other factors such as residual TGFβ ⁵⁰ in serum which may influence differentiation, we applied the TGF-β1 receptor blocker SB431542 as a control for TGF-β1 treatment. As shown in Figure 7C, the increased protein expression of α-SMA induced by TGF-β1 was significantly reduced by TRPM7 shRNA, which is consistent with the immunostaining result in human fibroblasts (Figure 7A-B).

Role of TRPM7 in TGF-β1 induced fibroblast proliferation

We next investigated the role of TRPM7 in proliferation of fibroblasts. Human cardiac fibroblasts treated with scramble or TRPM7-shRNA were exposed to TGF-β1 (10 ng/ml). As shown in Figure 7D, TRPM7-shRNA significantly decreased the number of fibroblasts, suggesting that TRPM7 is also involved in cardiac fibroblast proliferation caused by TGF-β1.

Regulation of TRPM7 by TGF-β1

As TGF-β1 is the most potent fibrogenesis stimulator, we studied whether expression of TRPM7 is modulated by TGF-β1. In this experiment, fibroblasts were treated with TGF-β1 or the TGF-β1 receptor blocker SB431542. TRPM7 currents were recorded after the human atrial fibroblasts were treated for 24 hr. As shown in Figure 8A-B, TRPM7 current density was increased by about 80% by TGF-β1. Ca²⁺-influx in TGF-β1 treated fibroblasts was significantly larger than that of control group (Online Figure VIII). Consistent with the increased current amplitude, TRPM7 expression at mRNA level assessed by qPCR using mouse cardiac fibroblasts was increased by 2.8 fold. Although other Ca²⁺-permeable channel genes were also up-regulated by TGF-β1 (Online Figure IX), TRPM7 is the only functional channel which can be measured by current recording in human atrial fibroblasts (Figure 2-3) and in mouse fibroblasts before and after after TGF-β1 treatment (data not shown). Moreover, knocking down the abundantly expressed TRPC1 did not alter Ca²⁺ influx or differentiation of human atrial fibroblasts (Online Figure IV). Therefore, it is likely that TRPM7 plays a major role in mediating Ca²⁺ in atrial fibroblasts under normal conditions and upon TGF-β stimulation. To further understand the correlation between the changes of TRPM7 induced by TGFβ1 and fibroblast differentiation, we tested TRPM7 protein and α-SMA expression levels by using mouse cardiac fibroblasts. Figure 8C-D shows that TGF-β1 increased TRPM7 and α-SMA protein expression by 3.8- and 2.1-fold, respectively. The mRNA level of collagen (type I) as well as collagen production measured in human atrial fibroblasts was markedly increased by TGF-β1 (Fig. 8E-F). These results further suggest that up-regulation of TRPM7 underlies the Ca²⁺ signaling mechanism for TGF-β1 induced fibroblast differentiation and collagen production, and that TRPM7 may play a role in TGF-β1 induced atrial fibrogenesis.

TRPM7 underlies the molecular mechanism of Ca²⁺ signaling in human atrial fibroblasts

Ca²⁺ signals are essential for various cellular functions. Although the importance of fibrosis in various cardiac diseases has been well appreciated, the Ca²⁺ signaling mechanisms in cardiac fibrogenesis are not known. Here we demonstrate that TRPM7 is abundantly expressed in human atrial fibroblasts, and is the major Ca²⁺-permeable channel which is responsible for Ca²⁺ influx in human atrial fibroblasts.

TRPM7 is a Ca²⁺-permeable cation channel ^{36, 37}. It is constitutively active and brings Ca²⁺ into cells under physiological conditions. High intracellular Mg²⁺ concentration blocks TRPM7. However, TRPM7 channels show ~50% of their maximum activity at physiological Mg²⁺ levels (0.5–1 mM) ^{36, 51}. Extracellular low pH potentiates TRPM7 inward current ⁴², and oxidative stress activates TRPM7 ⁴⁰. Membrane stretch ⁵² and shear stress ³⁸ have been reported to activate TRPM7. All of these characteristics of TRPM7 suggest that TRPM7 plays a role under pathological conditions, such as ischemia or oxidative stress, which may initiate the cardiac fibrogenesis cascade.

Our results indicate that TRPM7 is encoded by TRPM7 and that TRPM7 is the major functional Ca²⁺-permeable channel in human atrial fibroblasts. Although TRPV2, TRPV4, TRPC1 and TRPC6 were detectable by RT-PCR, current recording results indicate that there were no functional TRPV2, TRPV4 or TRPC6 channels in the fibroblasts, making them unlikely to contribute to Ca²⁺ signaling in the fibroblasts. TRPV4 is sensitive to stretch ⁵³, and can be specifically activated by 4α-PDD. We found that in both freshly isolated and cultured fibroblasts, 4α-PDD was unable to activate TRPV4-like currents (Figure 3). TRPC6 is activated by G_q-linked receptor stimulation, DAG or OAG, or by direct application of GTPγS in the pipette solution. We could not activate TRPC6-like current by LPA, which activates the EDG7 receptor ⁵⁴, or by including GTPγS in the pipette solution (Online Figure III). These results indicate that TRPC6 is not functionally expressed in human atrial fibroblasts. A previous study reported that TRPC6 was activated by C-type natriuretic peptide through PLC activation or directly by OAG in rat cardiac fibroblasts ⁴⁸. We do not know the reason for this discrepancy, but human and rat cardiac fibroblasts may express TRPC6 differently.

Store-depletion activated Ca²⁺ current (I_CRAC) contributes to Ca²⁺ influx and plays an important role in cellular functions in a variety of cells ^{55, 56}. In human atrial fibroblasts, we could not detect I_CRAC currents, and did not find that store-depletion could induce Ca²⁺ influx (Online Figure V) in human atrial fibroblasts. I_CRAC, TRPV2, TRPV4, TRPC1 and TRPC6 currents were not present in AF fibroblasts. Even though we found that TGF-β1 up-regulates mRNA expression of TRPM7, TRPV2, TRPC1, TRPC3, TRPC4 and TRPC6 (Online Figure IX), TRPM7 current is the only functional current recorded in the fibroblasts treated with TGF-β1, suggesting that TRPM7 is the major channel mediating Ca²⁺ entry in atrial fibroblasts. When TRPM7 was knocked down, Ca²⁺ influx was substantially decreased in TRPM7-shRNA treated atrial fibroblasts in comparison with Scr-shRNA treated atrial fibroblasts, indicating that TRPM7 is needed for Ca²⁺ entry in fibroblasts. Taken together, our results indicate that TRPM7 is the major Ca²⁺-permeable channel which is responsible for Ca2+ influx in human atrial fibroblasts.

Up-regulation of TRPM7 may underlie TGF-β1 induced fibroblast differentiation in AF patients

We found that TRPM7 is markedly up-regulated in the fibroblasts from AF patients. Both current density and Ca²⁺ influx mediated by TRPM7 from AF patients were significantly larger than those from NSR patients (Figure 4). Due to the limited amount of tissue and numbers of human atrial fibroblasts, we could not compare the TRPM7 expression in AF and NSR patients by measuring protein levels using Western Blot. Since the only differences between AF patients and NSR patients is the presence of atrial fibrillation versus normal sinus rhythm (Online Table I), it appears that atrial fibrillation is the only variable that correlates with the up-regulated TRPM7. To understand the causal importance of TRPM7 in AF, we investigated whether TRPM7 influences fibroblast differentiation. We found that fibroblasts isolated from AF patients are more prone to differentiation under culture conditions even without TGF-β1 treatment. This inherent tendency to differentiation could be due to the increased expression level of TRPM7 when the fibroblasts were exposed to the residual amount of TGFβ in the culture media containing serum ⁵⁰. Indeed, knocking-down TRPM7 by TRPM7-shRNA not only markedly reduced basal differentiation of AF fibroblasts (Fig. 6C-D), but also substantially decreased the number of differentiated myofibroblasts and the expression level of α-SMA induced by TGF-β1, indicating that TRPM7 is necessary for fibroblast differentiation. As the myofibroblasts are the primary cells that synthesize ECM and cytokines, which in turn stimulate the fibrogenesis cascade, the requirement of TRPM7 for myofibroblast differentiation indicates that TRPM7 mediated Ca²⁺ signals play a pivotal role in fibrosis formation.

Fibrogenesis can be triggered by various stimuli, including mechanical stretch, oxidative stress, hormones, myocardial injury, and autocrine-paracrine mediators ^{14, 16}, among which TGF-β1 is the most potent stimulator in both in vivo and in vitro conditions. Interestingly, we found that TRPM7 expression was significantly increased by TGF-β1 in cultured fibroblasts (Figure 8). The increased TRPM7 expression level induced by TGF-β1 parallels the fibroblast differentiation as assessed by α-SMA expression and collagen production (Figure 8C-F). These results indicate that TRPM7 is indispensable for TGF-β1 induced differentiation, and that the up-regulated TRPM7 may synergize with TGFβ1 in inducing fibroblast differentiation. It is possible that the up-regulation of TRPM7 by TGF-β1 in cultured fibroblasts represents the major mechanism by which TRPM7 is up-regulated in AF patients in vivo. Taken together, our results suggest the following model: In human atria, TGF-β1 is activated by mechanical stretch, oxidative stress, or other stimuli. TGF-β1 up-regulates TRPM7, which in turn brings more Ca²⁺ for TGF-β1 to induce differentiation. The differentiated myofibroblasts synthesize ECM and cytokines, and excessive accumulation of ECM forms fibrosis. Cytokines can in turn stimulate TGF-β1 and cause up-regulation of TRPM7, leading to perpetuation of the fibrogenesis process.

Potential mechanism by which Ca²⁺signals are involved in TGF-β1 induced fibrosis

Atrial fibrosis plays an important role in the pathology of AF. Although three interrelated pathways including the renin-angiotensin system, TGF-β1 and the oxidative stress pathways have been shown to be involved in fibrosis ^57-63, many fundamental aspects of fibrogenesis are still to be established ^{6, 57}. Our results indicate that TRPM7-mediated Ca²⁺ signal is another fundamental factor contributing to atrial fibrogenesis cascade.

How does the Ca²⁺ signal contribute to fibrosis formation? There are several possibilities. First, Ca²⁺-dependent but TGF-β1-independent pathways may control fibrotic gene expression ^64-67. Second, Ca²⁺ signals may influence TGF-β1 via the Smad-independent pathway ^68-70. Third, TGF-β1 can induce Ca²⁺ entry ^{71, 72} or integrate with reactive oxygen species (ROS) associated Ca²⁺ influx and therefore influence ECM protein synthesis ⁷³ through a ROS activated calcineurin pathway. As TRPM7 can be activated by ROS ⁴⁰, it is possible that the ROS activated calcineurin pathway may have contributed to the up-regulation of TRPM7 by TGF-β1 in human atrial fibroblasts. Fourth, TRPM7-mediated Ca²⁺ signals may activate the calcineurin pathway and produce synergistic effect with TGF-β1 (Smad-dependent pathway) in regulating gene expression. Further studies are required to understand how TRPM7 mediated Ca²⁺ signals contribute to TGF-β1 induced fibrogenesis. Nevertheless, our findings that TRPM7 is the molecular basis for the major Ca²⁺ entry channel in cardiac fibroblasts provide a necessary tool to unravel the mechanisms by which Ca²⁺ signals contribute to fibrogenesis.

Potential significance

Atrial fibrosis can be triggered by various factors ¹³ via different pathways ⁵⁷. Blocking one pathway may not be able to attenuate or eliminate fibrosis given the complex pathological conditions. The TRPM7-mediated Ca²⁺ signals which are necessary for TGF-β1 induced fibrogenesis may serve as a common pathway in the fibrogenesis cascade. If so, inhibition of Ca²⁺ entry through TRPM7 may attenuate cardiac fibrosis regardless of the original stimuli.

TRPM7 has been shown to be essential for cell viability ^{74, 75} and is responsible for Ca²⁺ overload induced cell death under anoxia conditions ⁴⁰. Although previous studies suggested that TRPM7 is involved in Mg²⁺ homeostasis ^{74, 76}, knocking out of the TRPM7 gene disrupts embryonic development but does not alter Mg²⁺ homeostasis ³⁹, indicating that Ca²⁺ permeation may confer major channel functions of TRPM7. The contribution of the TRPM7-mediated Ca²⁺ signal in human atrial fibrogenesis shown in this study is the first demonstration of an important role of this unique channel in human diseases.

Conclusions

In this study, we establish that the Ca²⁺-permeable cation channel TRPM7 underlies Ca²⁺ signaling mechanisms in the cardiac fibrogenesis, making TRPM7 the first known Ca²⁺-permeable channel involved in the genesis of cardiac fibrosis. More importantly, our results demonstrate that TRPM7 plays an essential role in fibrogenesis in human AF. Our study opens a new avenue to explore the potential roles of Ca²⁺ signaling in fibrogenesis, which will lead to better understanding of the mechanism of cardiac fibrogenesis, and could ultimately results in development of more effective approaches for treatment of AF.