Ciclopirox Olamine Induces Ferritinophagy and Ameliorates Disease Progression in Polycystic Kidney Disease

Despite the recent launch of Tolvaptan, the search for safer polycystic kidney disease (PKD) drugs continues. Ciclopirox, as the free acid (CPX) and its olamine salt (CPX-O) , are contained in number of commercially available topical antifungal agents. CPX is reported in the literature to possess anticancer activity in number of solid tumor cancers and hematological malignancy by several proposed mechanisms of action including chelation of iron and inhibition of iron dependent enzymes. Here, we show that CPX-O inhibited in vitro cystogenesis of primary human cyst epithelial cells cultured in 3D collagen matrix. To determine if CPX-O inhibits PKD progression, we treated PKD mice with a low dose of 10 mg/kg CPX-O by daily intraperitoneal injections from day 21 to day 49 postpartum. CPX-O reduced the kidney to body weight ratio of the PKD mice. This was associated with decreased cell proliferation decrease cystic area and improved renal function. We found that ferritin levels were significantly elevated in cystic kidneys of PKD mice, and that CPX-O treatment reduced renal ferritin levels and increased ferritinophagy marker, NCOA4. Our data suggest that CPXO dose dependently induces ferritin degradation via ferritinophagy which is associated with decreased cyst growth and disease progression in PKD mice. Most importantly these data indicate that CPX-O, a drug used to treat skin Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 February 2020 doi:10.20944/preprints202002.0416.v1


Introduction
Polycystic kidney disease (PKD) is a genetic disease characterized by multiple cysts in the kidneys which enlarge overtime and lead to end stage renal disease.
The incidence of autosomal dominant polycystic kidney disease (ADPKD) is 1 in every 1000 births. About 85% of ADPKD cases have mutations in polycystin 1 (PC1) and the remaining 15% in polycystin 2 (PC2) (proteins encoded by PKD1 and PKD2 respectively) 1,2 . Both PC1 and PC2 have been found to be localized to cilia. PC1 and PC2 function in the cilia and appear to be important in the regulation of intracellular Ca 2+ .Decreased intracellular Ca 2+ levels in ADPKD cells in combination with activation of adenylate cyclase and accumulation of cAMP lead to increased proliferation and fluid secretion and contribute to enlargement of cysts [3][4][5][6] . Many signaling pathways have been shown to be deregulated in response to loss of function mutations in PKD 2,3,7,8 .
We identified a drug Ciclopirox (CPX) or its olamine salt (CPX-O) as a drug candidate that could influence cyst progression as this drug has been shown to alter many pathways that are also affected in PKD such as the Notch signaling pathway, the mTOR pathway, the MEK-ERK pathway and the Wnt signaling pathway to name a few [9][10][11][12][13] . CPX is an antifungal, topical solution used in a variety of formulations such as gels, creams, shampoos for treatment of cutaneous fungal infections 14 . However, in last decade, the wide use of CPX in other diseases have been recognized [15][16][17] . Furthermore, the safe therapeutic response of CPX has been tested in various animals including pigs and dogs 18 .
The preclinical studies have shown that systemic administration of CPX in hematologic malignancies resulted in antitumor effects 19 . The proposed mechanisms of CPX include chelation of polyvalent metal cations, such Fe3+ by which it can inhibit iron dependent enzymes, such as ribonucleotide reductase and components of hypoxia, Wnt, and Notch signaling 10,20,21 . CPX is also involved in ferritin degradation 22 and protection against ferroptosis [23][24][25] . Ferritin is a hetero-polymeric molecule composed of 24 heavy and light subunits which can store excess iron in cells of up to 4500 iron atoms 26 . Ferritin is composed of varying proportions of two gene products, heavy ferritin (FTH1) and light ferritin (FTL). The FTH1 exhibits ferroxidase activity which converts toxic ferrous ions into ferric ions, while the light subunit facilitates iron storage [27][28][29] . Given the role of CPX as an iron chelator and the role in multiple pathways shown to be altered in PKD, we hypothesized that CPX-O ameliorates disease progression in PKD by mechanisms that involve iron-ferritin axis.
Our data indicate that CPX-O can reduce cyst formation in 3D collagen assays from primary cystic epithelial cells from ADPKD patients. CPX-O can also attenuate disease progression in a PKD mouse model. Mechanistically, CPX-O appeared to reduce the elevated ferritin expression in cystic epithelial cells by the process of ferritinophagy both in vivo and in vitro.

CPX-O inhibits cyst formation of ADPKD cells in 3D cyst assays
To first validate whether CPX-O may be a drug of choice for PKD in the clinics, and whether further mouse studies are required to validate it, we performed 3D cyst assays using primary ADPKD renal epithelial cells 30

Protection mediated by CPX-O does not involve Notch pathway inhibition
We have previously shown that the Notch3 signaling pathway was activated in human ADPKD cyst lining cells 31

CPX-O reduces cell proliferation and fibrosis in PKD
To determine whether CPX-O mediates affect on cell proliferation, kidney sections were stained for Ki67 antigen and couter stained with hematoxylin to count total cells. Ki67 positive cells were clearly seen in abundance in the cystic epithelia (Fig. 4a, arrow). Figure 4B represents quantitative analysis where Ki67 labeling is presented as percent positive cells per section (n=17each group), which represented three mice in each group. In addition, Mason trichrome staining and alpha smooth muscle actin labeling was less intense in CPX-O treated PKD sections compared to vehicle treated PKD sections, suggesting a decrease in fibrosis (supplementary fig. 2).

CPX-O induces ferritinophagy to normalize ferritin expression in PKD
Ferritin degradation via autophagy has been a proposed mechanism of action of CPX-O 11,22 . Autophagy is a process that results in the formation of autophagosomes in which double membraned vesicular structures sequester cytoplasmic components and fuse with lysosomes to form autolysosomes. In these structures, engulfed cargo is broken down by lysosome-derived acid hydrolases. When nuclear receptor coactivator 4 (NCOA 4), a cargo receptor, binds to ferritin, it targets ferritin degradation, a process that is termed" ferritinophagy" [34][35][36][37][38] . We determined whether ferritin reduction in CPX-O treated PKD mice involved autophagy and more specifically ferritinophagy. WB analysis in kidney lysates from mice revealed that an autophagy marker, LC3B was downregulated in vehicle-treated PKD samples (n=4) relative to vehicle-treated WT samples (n=3) (Fig. 6A). Interestingly, CPX-O was able to clearly upregulate LC3B-II levels in CPX-O treated PKD samples compared to vehicle-treated PKD samples, whereas CPX-O did not appear have a difference in the WT kidneys.
These data were similar as NCOA4 expression, which was downregulated in PKD indicating reduced ferritin degradation and reversal of this phenomenon by CPX-O and hence upregulated NCOA4. This process was also observed in wildtype CPX-O treated mouse kidney lysates however, the ferritin staining was weaker compared to kidneys from PKD mice.
To determine whether the induced ferritinophagy is a direct effect of CPX-O, we The nuclear labeling in primary cells appeared to be only a property of cells in culture because such effect was not seen in human or mouse kidneys. These data indicate that ferritinophagy, as demonstrated by the loss of ferritin, is accompanied by an increase in NCOA4.

Discussion
To our knowledge, we report for the first time that ferritin is upregulated in renal cells in PKD, and CPX-O normalizes the ferritin levels and protects against cyst progression in ADPKD.
In the kidneys, proximal tubules are the major source of heavy chain ferritin (FTH1) expression 39 . The current study with human primary cells shows ferritin upregulation in cystic cells which are mainly of collecting duct origin in ADPKD kidneys. Also DBA positive cells in PKD kidneys, stained positive for ferritin, indicating that under pathologic circumstances, ferritin is increased in the cystic cells, similar to that reported in cancer and other pathologic diseases [40][41][42][43][44] .
Conditional knock down of proximal tubule specific FTH1 in mice worsened acute kidney injury and was associated with increased apoptosis and significant mortality. The study indicated a protective role of FTH1 in proximal tubules in AKI 39 . On the contrary, deletion of myeloid FTH1 was protective against lipopolysaccharide induced endotoxemia in mice. In this model compensatory increased in circulating FTL was reported 45 . In addition, there are reports on iron independent roles of ferritin 44 . Ferritin levels in serum correlate with the body's iron stores, however under inflammatory conditions such as cancer or autoimmune diseases, elevated ferritin levels are reported which do not correlate with the body's iron levels 43 . In breast cancer cells, the binding and uptake of ferritin was observed along with increased cell growth independent of the iron status of ferritin 44 . Our data suggest that in PKD ferritin upregulation is a pathogenic response, whether it is iron dependent or independent would need further studies.
Ferritin treatment on endothelial cells promoted activity of Erk and AKT signaling promoting pro-angiogenic effects 41 . These pro-survival pathways are also activated in ADPKD cells and thus it is tempting to speculate that ferritin accumulation may play a role in elevating these pathways as well as other pathways that are activated in PKD 3 . In our studies, the Notch 3 pathway is activated in PKD tissues which correlated with the ferritin increase, but CPX-O (10mg/ml body weight) mediated ferritin reduction did not correlate with the Notch 3 pathway which indicates that in kidney, CPX-O does not alter Notch 3 signaling pathway at the low dose used in our study.
In recent years, CPX-O has emerged as an important anticancer agent and is currently in clinical trials for advanced hematological malignancies 19 . The protective effects of CPX-O have been attributed to its potential to chelate iron and thus effect iron dependent pathways and enzymes, which may potentially lead to inhibition of cell proliferation and induction of cell death in tumors 13 . Iron is an essential element in the body and also regulates the heme pathway.

Unbound intracellular iron generates reactive oxygen species (ROS) through
Fenton chemistry leading to DNA breaks, lipid peroxidation, and cellular damage.
Thus, a tight regulation of iron metabolism is required to maintain cellular homeostasis. However it is known that CPX-O is a mild iron chelator and its use in small doses does not appear to affect the heme group, instead it was recently suggested that CPX can stabilize heme groups instead of competing with them for metal chelation 46 .
In our studies we find that ADPKD cells are more sensitive to CPX-O than the NHK cells. This was shown by the dose response curve in supplementary data where NHK cells resisted cell death upto 10µM concentration, whereas ADPKD cells responded to 1µM or less. In 3D collagen assays, only 0.2 to 0.5µM was enough to significantly inhibit cyst growth. In mouse studies we used only 10mg/g body weight dose which was sufficient to reduce cystogenesis and normalize kidney function in PKD mice. Our studies suggest that low doses of CPX-O are enough for the cystic cells to respond.
The autophagic degradation of ferritin, an iron storage protein has been reported to be a key mechanism maintaining iron homeostasis 38,47 . We show that ferritin expression is increased in cyst lining epithelial cells and CPX-O can induce ferritinophagy in these cells. The process of ferritinophagy involves release free iron and resulting in ferroptosis (iron induced cell death), which could be detrimental for cells, a possibility is that CPX-O provides dual protection in PKD by inducing ferritinophagy and protecting against ferroptosis as described previously 24 . Whether CPX-O results in protection against ferroptosis in PKD will require further investigation.
This study opens avenues for future studies regarding the role of iron metabolism in PKD. Available literature on extracellular ferritin studies shows that ADPKD patients have higher levels of hemoglobin compared to other forms of CKD patients. This has been thought to be a consequence of increased erythropoietin production by cystic cells 48,49 . Average ferritin levels of 100-800 ng/mL was associated with the best survival in PKD patients, whereas that of non-PKD patients was 500-800 ng/mL, showing that iron may play a role in PKD 50 .
Sequestration of intracellular iron was speculated as a mechanism for cyst reduction by NGAL (Neutrophil gelatinase-associated lipocalin 51  Wildtype littermates without a PKD mutation (Pkd1 +/+/ /Pkd2 +/+ ) were used as controls. The PKD1 RC/RC , PKD2 +/-(PKD) mice are mildly cystic at birth and the cysts progress with age such that there is an exponential growth of the cysts between age P15 and P60 after which cyst growth does not advance. We used these mice because the time period of cyst formation was moderate for the drug studies. Biosciences, Houston, TX) and manufacturer's instructions were followed.

CPX-O Treatments in mice:
Viability was set at 100% for the vehicle control and relative values were calculated for other doses.
In vitro 3D cyst assays: In vitro cyst assays were performed as described 6 30,55 .
Briefly, primary cultures of ADPKD cells were suspended in media containing type 1 collagen (PureCol, Advanced biomatrix, San Diego, CA) in a 96-well plate.
Immediately after adding collagen and cells (4 × 10 3 /100ul), 100ul of media with collagen and cells was pipetted into each well of the 96-well plate. The plate was incubated at 37 °C for 45 minutes to allow collagen to polymerize. Then, Protein concentration was measured using BCA protein assay (Bio-Rad, Hercules, CA). Whole cell lysates (50 to 100μg) were solubilized in 6X Laemmli sample buffer and heated to 95 °C for ten minutes and electrophoresed on 15% (for ferritin and LC3B) and 10% (for other proteins) polyacrylamide gels. Proteins were transferred to PVDF membranes. Ponceau staining was performed for each blot to determine protein transfer and imaged. Ponceau S (total protein was used to normalize the gels for protein loading 57 ). The immunoblots were blocked in 5% nonfat dry milk in PBST (PBS containing 0.1% Tween 20) for 1 hour at room temperature and then followed by PBS washes; the blots were incubated with appropriate dilutions of primary antibodies overnight. The blots were then washed and incubated with secondary antibodies (1:10,000 dilution in blocking solution) for 1 hour at room temperature. After subsequent washes in PBST, bound antibody was detected by chemiluminescence (Western Lightning Plus ECL, Perkin Elmer). Bands produced in the results were quantified using imageJ and normalized with ponceau S staining to confirm equal loading 57 . Data was presented as relative fold change.

Statistics
Data are expressed as mean ± SE. Statistical significance was measured by Student's unpaired T test for comparison between control and PKD groups. Oneway ANOVA was performed to compare more than two groups followed by Tukey HSD test. A P value of <0.05 was considered statistically significant.

Disclosure
The authors have no disclosures.   Western blot (WB) analysis for Notch3IC, Hes1 and RBPjk (Notch targets) shows increased expression of Notch targets in kidney lysates of vehicle (Veh)treated PKD mice, compared to Veh-treated WT mice. (C, D and E) Protein blots were quantified for N3IC (n=3), Hes1 (n=5) and RBPjk (n=4) using Ponceau S staining for total protein normalization. Statistical significance was determined using One-way ANOVA followed by Tukey's HSD test (*p<0.05).  quantification (right panel) shows increased ferritin in veh-treated PKD mice relative to WT mice and reduced ferritin in CPX-O treated PKD mice relative to veh-treated PKD mice (n=5-6/group). (C) WB for ferritin using lysates of primary cells from normal or ADPKD patients revealing increased ferritin in ADPKD (n=3 patients/group) (left). Quantification is shown in the right panel.
(D) Upper panels represent normal human kidney section as control and lower panel represents an ADPKD patient kidney sample with arrows indicating high ferritin intensity both in cyst epithelium and interstitium. (E, F, G and H) Colabelling of ferritin (green) with DBA (a collecting duct marker, red), DAPI (a nuclear stain, blue). Upper panels represents a PKD kidney with no ferritin antibody to serve as a control. Lower panels show ferritin in green. (H) Merged image from panels E, F and G showing ferritin co-localization with DBA (pink) in PKD sections only. One-way ANOVA followed by Tukey's HSD test and unpaired Student T-test were used for statistical analyses of graphs in (B) and (C)(*p<0.05, **p<0.01, ***p<0.001).