Calcium, TRPC channels, and regulation of the actin cytoskeleton in podocytes: towards a future of targeted therapies

A 16-year-old boy who presented with nephrotic syndrome since the age of 7, with an initial biopsy showing minimal change disease (MCD), and with a subsequent biopsy showing primary FSGS (focal segmental glomerulosclerosis) at age 9, presented with relapsing steroid-resistant nephrotic syndrome. The patient had preserved renal function with an average serum creatinine 0.5 mg/dl. His medications included lisinopril 10 mg daily, simvastatin 40 mg daily, and furosemide as needed, titrated to symptoms. Cyclosporine A, cytotoxic therapy with cyclophosphamide, mycophenolate mofetil, and rituximab had been added to steroids at different times during his adolescent years, without success in achieving remission. At age 16, while in partial remission on high-dose oral steroids, and during a subsequent attempt to slowly taper the steroids, he experienced recurrence of proteinuria with a urinary protein/creatinine ratio (UPC) of 12. Multiple attempts to achieve remission with high-dose steroids were unsuccessful. A few weeks later, he was admitted to the hospital with anasarca that required intravenous diuretics. A repeat kidney biopsy confirmed the diagnosis of primary podocytopathy or FSGS. What further therapeutic interventions may be available to this patient with steroid-resistant nephrotic syndrome?

Ca²⁺ is a vital second messenger in virtually every cell type [8]. It orchestrates a great variety of different cellular functions such as muscle contraction [9], fertilization [10], neurotransmitter release [11], cytoskeleton dynamics [12] and apoptosis [13, 14]. Ca²⁺ signaling relies on a low intracellular Ca²⁺ resting concentration [Ca²⁺]_i (0.1 μM), roughly four orders of magnitude below the extracellular Ca²⁺ concentration [Ca²⁺]_e (2 mM). A Ca²⁺ signal is induced by an increase of the [Ca²⁺]_i, whereby the major sources of Ca²⁺ are the extracellular space and intracellular Ca²⁺ stores such as the endoplasmic reticulum (ER) and the mitochondria [15]. Specialized ion channels, pumps, and cytosolic Ca²⁺ buffers constitute the cellular Ca²⁺ regulatory apparatus. Their well-coordinated interplay allows both the induction of intracellular Ca²⁺ signals (ON mechanisms) and the relaxation of the system towards its vital low resting [Ca²⁺]_i (OFF mechanisms) [16]. ON mechanisms are usually mediated by ion channels which, upon opening, allow Ca²⁺ to diffuse along its concentration gradient into the cytosol. Common examples include voltage-dependent Ca²⁺ channels (VDCCs) and intracellular ligand-gated Ca²⁺ release channels, such as the inositol 1,4,5-trisphosphate receptor (IP₃R) or the ryanodine receptor (RYR), predominantly located on the ER membrane [15, 16]. OFF mechanisms, by contrast, extrude Ca²⁺ from the cytosol, thus pumping Ca²⁺ against its concentration gradient either into the extracellular space or into intracellular Ca²⁺ stores. Important examples are the sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) and the plasma membrane Na⁺/Ca²⁺-exchanger (NCX) [15, 16]. Upon elevated [Ca²⁺]_i, Ca²⁺ binds to highly conserved Ca²⁺ binding domains, such as EF-hand domains [17]. By introducing its two-fold positive charge, Ca²⁺ induces a conformational change in these proteins, which can lead to an alteration of their function. Ca²⁺-sensing proteins are functionally highly diverse and range from transcription factors and phosphatases [8] to elements of the cytoskeleton [12] and proteins of the contractile apparatus in muscle cells [18]. A prolonged increase in [Ca²⁺]_i induces cell death [13, 14] and thus Ca²⁺ signals are required to be temporally limited. Furthermore, to be able to mediate the great variety of Ca²⁺-dependent cellular functions, Ca²⁺ signals are not necessarily binary signals, but can also encode information in a frequency-dependent manner (for example, as Ca²⁺ oscillations) [18]. Spatial heterogeneity is another important aspect to understand the high specificity of Ca²⁺ signals. Instead of looking at the cell as a single homogeneous system, it is necessary to identify so-called Ca²⁺ microdomains, sub-femto liter domains that accumulate important Ca²⁺ signaling elements. Such microdomains can be predetermined by structural features of the cell, such as dendritic spines in neurons [19] or FPs in podocytes [20], or their spatial extent can be limited by diffusion and the cytosolic Ca²⁺ buffering capacity. Acknowledgment of the importance of Ca²⁺ microdomains to our understanding of Ca²⁺ signaling in general is constantly increasing [21], which is reflected in a growing number of theoretical studies investigating the mathematical foundations of the complex non-linear dynamics of Ca²⁺ microdomains [22‐25].

Transient receptor potential (TRP) channels are a heterogeneous group of non-selective cationic channels, containing 28 human homologs, subdivided into seven subfamilies [26, 27]. Two members of the canonical TRP (TRPC) channel subfamily have been identified to play a major role in the Ca²⁺ homeostasis of podocytes, namely TRPC6 and TRPC5 [28] (Fig. 1). Human genetics has revealed TRPC6 as a vital component of the slit diaphragm [29], and further investigation into gain-of-function TRPC6 mutations ultimately led to the identification of TRPC5 as a critical inducible mediator of Ca²⁺-induced cytoskeletal remodeling in podocytes [28], an important event at the onset of all podocytopathies [30, 31].

The beneficial therapeutic effects of renin–angiotensin–aldosterone system (RAAS) inhibition in primary podocytopathies are well established [32]. However, molecular insights into the underlying mechanisms were lacking until the AT1R was identified as the upstream receptor signaling to both TRPC5 and TRPC6 in podocytes [28]. The Gq subunit of the G-protein-coupled AT1R activates phospholipase C (PLC), which in turn controls TRPC5 /TRPC6 gating. While the detailed mechanisms for differential and specific activation of each channel have not yet been fully elucidated, the activation of both channels leads to Ca²⁺ influx into podocytes, with distinct cellular responses [28].

Tian et al. showed that while TRPC6 activates RhoA, inducing a stationary, homeostatic podocyte phenotype, TRPC5 activates Rac1 inducing a disease-associated motile phenotype: in other words, triggering a cellular program that centers on Rac1 to mediate maladaptive cytoskeletal remodeling leading to FPE [28, 30]. Indeed, prior to the work revealing TRPC5/Rac1 and TRPC6/RhoA coupling, the small Rho GTPases (RhoA, Rac1, and Cdc42) had been associated with different aspects of cell motility and cytoskeleton dynamics in podocytes [31], but they had not been linked to the TRPC channels or to Ca²⁺ signaling.

However, how can individual TRPC channels mediate distinct cellular events? A unifying explanation can be found in the existence of Ca²⁺ microdomains in podocytes, to which each TRPC channel introduces small amounts of Ca²⁺, locally increasing [Ca²⁺]_i, sufficient to activate a localized effector [28]. Thus, both TRPC5 and TRPC6 are essential for a functioning podocyte cytoskeleton: gene silencing of TRPC6 induces the activation of Rac1, the loss of stress fibers and increased podocyte motility, in summary a motile phenotype. Constitutive activation of RhoA rescues the phenotype [28]. Interestingly, similar to gain-of-function mutations in humans, overexpression of TRPC6 in podocytes in vitro also leads to loss of stress fibers in podocytes [33] and proteinuria in mice [34], indicating the necessity of a well-balanced TRPC6 activity to maintain a healthy podocyte cytoskeleton (Fig. 1). Conversely, gene silencing of TRPC5 induces stress fiber formation and RhoA activation and decreases podocyte motility, and thus leads to a stationary phenotype. In this case, constitutive activation of Rac1 rescues the phenotype [28]. In summary, these findings suggested that instead of the activity of one particular pathway in podocytes, it is the balanced activity of both TRPC6/RhoA and TRPC5/Rac1 that is critically important. Subsequent work in vivo corroborated these findings and further showed that, in the setting of early proteinuric disease, TRPC5 is induced to activate Rac1 leading to FPE. Most importantly, this could be blocked by a small molecule inhibitor, ML204, in a dose-dependent fashion. This work revealed a new, ion channel-targeted approach to anti-proteinuric therapy [30].

The Ca²⁺-dependent phosphatase calcineurin is typically known for activating the transcription factor nuclear factor of activated T-cell (NFAT) in T-cells, upregulating interleukin 2 (IL-2), and thus inducing the T-cell response [35]. However, calcineurin has also been shown to have a distinct role in podocytes. Calcineurin activation leads to cathepsin L-mediated cleavage of the vital, actin-associated protein synaptopodin [36], which plays an essential role in podocyte cytoskeletal homeostasis [37]. Activation of calcineurin induces synaptopodin cleavage and proteinuria, while the calcineurin inhibitor cyclosporine A (CsA) prevents it [38]. This finding challenged the idea that the treatment of proteinuric kidney diseases with CsA is based on immunomodulation, and emphasized instead the importance of an intact podocyte cytoskeleton for the glomerular filtration barrier [3], revealing the podocyte as the therapeutic target of choice for glomerular diseases.

The fact that FSGS-causing TRPC6 gain-of-function mutations activate calcineurin-NFAT-dependent gene transcription [39] further points towards an important role of TRPC channels and calcineurin signaling in podocytes. In cardiac hypertrophy, the calcineurin-NFAT pathway induces the upregulation of TRPC6 transcription [40] and, as just recently proposed, a similar mechanism appears to exist in podocytes [41]. Perhaps this mechanism constitutes an auto-regulatory positive feedback loop, eventually increasing the presence of TRPC6 channel density on the plasma membrane. Further investigations need to be directed towards the understanding of this mechanism to elucidate the full scope of TRPC channel-induced podocytopathies.

The hallmark of primary podocytopathies is proteinuria, a direct consequence of filter barrier damage. Functionally, podocytes are critical for maintaining an intact filtration barrier [3]. Importantly, podocytes are terminally differentiated cells, quite similar to neurons, and therefore their capacity for regeneration is limited. Podocytes have developed potent “adaptive” mechanisms to cope with acute and chronic stress and injury. Also similar to neurons, podocytes have exquisite mechanisms for the tight control of Ca²⁺ homeostasis, because the consequence of excess Ca²⁺ is cytoskeletal collapse [30, 52‐55] without the possibility of a suitable replacement. In this scenario, FPE may be seen as an initial (futile) attempt to maintain a (filtration) barrier in the face of injury. Clinically, this manifests as proteinuria, and precedes the presence of histological sclerotic lesions, which are the result of podocyte death and maladaptive proliferation of parietal epithelial cells on the denuded GBM [59].

The identification of TRPC5 and TRPC6 as upstream regulators of Ca²⁺-mediated cytoskeletal rearrangements in podocytes, and the notion of maintaining a proper balance between homeostatic TRPC6/RhoA versus inducible TRPC5/Rac1 activity constitutes a novel paradigm for understanding filter barrier damage in proteinuric diseases. Together with the development of improved TRPC channel inhibitors, these findings constitute a new approach to the unmet need of effective treatment options for primary podocytopathies such as MCD, FSGS, and diabetic kidney disease, as well as other acquired proteinuric kidney diseases. Furthermore, new drugs interacting with downstream targets of calcineurin, most importantly synaptopodin, could constitute a treatment option that is, due to an optimized side effect spectrum, more attractive than the currently used CsA (Fig. 2).

Patients with initial MCD diagnoses, resistant to steroid treatment, with subsequent biopsies showing FSGS, as described in the case report at the beginning of this review, appear frequently in the renal clinic. Following the here-proposed mechanistic understanding of the pathogenesis of primary podocytopathies, the seemingly distinct entities we call MCD and FSGS may not necessarily be two different diseases, but rather the transition from an initial state of a collapsed cytoskeleton (MCD) to a decompensated state of permanent podocyte loss and replacement with scar (FSGS). This new, molecularly-defined understanding of the mechanisms underlying primary podocytopathies also brings into crisper focus the therapeutic potential of targeting the podocyte cytoskeleton.

With more than 6,000 new pediatric patients suffering from treatment-resistant nephrotic syndrome in the US each year alone [60], the unmet needs of patients with treatment-resistant nephrotic syndrome is substantial, and calls for novel, podocyte-specific therapies. While first results are promising, further studies are necessary to advance towards a future when clinicians will have at their disposal an array of molecular targeted, precision medicine strategies for primary podocytopathies.