Acute and Chronic Iron Overloading Differentially Modulates the Expression of Cellular Iron-homeostatic Molecules in Normal Rat Kidney

Abstract

Little is known about the renal responses to acute iron overloading. This study measured the renal tubular expression of transferrin receptor-1 (TfR1), cubilin/megalin receptors, hepcidin, ferroportin, and ferritin chains following subacute intoxication of 40 male Wistar rats with a single oral dose of ferrous iron (300 mg/kg). The animals were randomly subdivided into 4 equal subgroups at the time of necropsy (1, 2, 4, and 8 hr). The results were compared with the controls (n=15) and with the chronic group (n=15), which received iron for 4 weeks (75 mg/kg/day; 5 days/week). Although both toxicity models inhibited TfR1, they upregulated the cubilin/megalin receptors and hepcidin, and triggered iron deposition in tubular cells. The ferritin heavy-chain and ferroportin were downregulated in the 2-hr and 4-hr acute subgroups, whereas chronic toxicity promoted their expression, compared with controls. Moreover, the 4-hr and 8-hr subgroups had higher intracellular Fe⁺² and marked cell apoptosis compared with the chronic group. In conclusion, the kidney appears to sustain iron reabsorption in both intoxication models. However, the cellular iron storage and exporter proteins were differentially expressed in both models, and their inhibition post-acute toxicity might contribute toward the intracellular accumulation of Fe⁺², oxidative stress, and ferroptosis.

Introduction

Iron is an essential trace element involved in a diversity of cellular functions.^1–3 Because an active excretory mechanism is absent, iron balance is maintained through the recycling of endogenous iron by the reticuloendothelial system and hepatic tissue together with restricting the amount of intestinal uptake.⁴ Iron homeostasis is regulated by the systemic iron-regulatory hormone, hepcidin, that is predominantly produced by hepatocytes in response to an increase in plasma iron levels.⁵ Hepcidin interacts with its receptor, ferroportin (FPN), which is the only known cellular iron exporter protein. The binding of hepcidin to FPN induces the degradation of the latter, thus, decreasing plasma iron levels through the inhibition of iron release from duodenal enterocytes, macrophages, and hepatocytes.^6,7

Iron circulates in the ferric (Fe³⁺) form bound to its carrier protein transferrin (Tf).⁸ The cellular uptake of Tf-bound iron (TBI) occurs by transferrin receptor-1 (TfR1)-mediated endocytosis followed by the release of the metal intracellularly in the catalyzed form, ferrous (Fe²⁺).^1,4,8 Iron is then stored bound to cytosolic ferritin, an acute phase protein composed of 24 heavy (H)- and light (L)-chains. Ferritin H-chain oxidizes the hazardous Fe²⁺ into Fe³⁺ following the nucleation of iron by the L-chain.^9–12 However, if intracellular Fe²⁺ surpassed the binding capacity of cellular ferritin, it could promote cell death by generating highly reactive oxygen species (ROS) through the Fenton’s reaction, a process known as ferroptosis.^13–15

The renal tubules reabsorb >99% of daily filtrated iron, and the cellular iron-regulatory proteins are expressed in renal tissues, advocating the importance of kidney in systemic iron homeostasis.^16–19 In this context, the transcytosis of TBI from the luminal border of renal tubules is mediated through TfR1.^16,20 Other apical receptors, cubilin and megalin, are also involved in the uptake of TBI in proximal tubules (PT).^21,22 In addition, both ferritin chains are synthesized by renal tubular cells, and the PT cells have been suggested as a major source of blood ferritin.⁹ Hepcidin and FPN are also expressed by renal tubules and are mainly localized basolaterally, suggesting that both proteins regulate iron transportation back into circulation.^18,23,24

Furthermore, renal diseases and iron dyshomeostasis are interrelated, and the occurrence of one of them could predispose to the other. Clinical studies have demonstrated that anemia induced by chronic kidney diseases (CKD) was characterized by dysregulations in iron metabolism, abnormal hepcidin levels together with aberrant iron deposition in renal tissue.^25–27 Experimental studies have also reported pathological alterations in the expression of several iron-regulatory proteins together with the accumulation of iron in renal tissues obtained from CKD, acute kidney injury (AKI), and chronic iron overloading.^28–32

Nevertheless, little is currently known regarding the effects of acute iron toxicity on the expression of iron-homeostatic proteins in renal tissues.^12,27,33 This study, therefore, measured the effects of acute sublethal iron overloading on the expression of iron-regulatory proteins together with several markers of oxidative stress and ferroptosis in renal tissues, and the results were compared with those obtained from controls and following chronic iron overloading.

Materials and Methods

Study Design and Iron Overloading Protocols

A total of 70 male Wistar rats of 12 weeks of age and 200–250 gm each were housed in a controlled temperature of 22–24C and 12-hr dark/light cycle. The animals were nourished with standard laboratory pellet diet and water ad libitum, and received human care throughout the study. The rats were divided randomly following housing for a week for acclimation into negative control (n=15), acute (n=40), and chronic (n=15) groups. All experimental protocols were in accordance with the European Union (EU) Directive 2010/63/EU for animal experiments and were approved by the Committee for the Care and Use of Laboratory Animals in Umm Al-Qura University.

Iron was freshly prepared by dissolving every 10 iron sulphate tablets (Feromin; Riyadh Pharma; Riyadh, Saudi Arabia) containing 600 mg of ferrous iron in 8 ml of sterile saline to prepare a final iron concentration of 75 mg/ml. Following food deprivation for 8 hr, the acute group received a single oral dose of 300 mg/kg of iron to induce acute sublethal iron overload as higher doses were associated with significant morbidity and/or mortality.³⁴ The animals were then allowed to freely access food and water, and were later subdivided randomly into 4 subgroups (n=10/group) according to the time of necropsy at 1 (A-1H), 2 (A-2H), 4 (A-4H), and 8 (A-8H) hr. The negative controls that were also deprived of food for 8 hr received normal saline by oral gavage, and were euthanized with the A-8H group. The chronic group received freshly prepared ferrous iron (75 mg/kg/day) by oral gavage for a total duration of 4 weeks (5 days/week),³⁵ and euthanasia was performed 8 hr after the last dose of iron.

Types of Samples

Euthanasia was done following anesthesia by diethyl ether (Sigma-Aldrich Co.; St. Louis, MO) and the collection of 3 ml of venous blood from each animal in a plain tube. Serum was stored in −20C until used. Both kidneys were collected from each animal, cut in halves, and a portion from each kidney was processed by a conventional method for histology experiments. Another piece weighing 25 mg from each kidney was immediately processed for total protein extraction using 2 ml cold RIPA lysis buffer containing protease inhibitors (Santa-Cruz Biotechnology Inc.; Santa Cruz, CA). The concentrations of total proteins were measured on Qubit Fluorometer (Thermo Fisher Scientific; San Diego, CA), and each sample was stored in −20C following dilution with normal sterile saline for a final concentration of 500 µg/ml.

The remaining renal tissues were immersed in 5 ml of RNALater (Thermo Fisher Scientific) and stored in −80C. Total RNA was then extracted using Purelink RNA mini kit (Thermo Fisher Scientific) according to the manufacturer’s instructions and following homogenization in sterile plastic tubes containing beads and a tissue raptor (Omni International; Kennesaw, GA). The quality of RNA was assessed on a BioSpec-nano machine (Shimadzu Corporation; Tokyo, Japan) and typically had an A260/A280 ratio of 1.7 to 1.9. The quantities of total RNA samples were measured on Qubit Fluorometer and then stored at −80C until used.

Biochemical Studies

Serum creatinine, blood urea nitrogen, iron, ferritin, and total iron binding capacity (TIBC) in addition to iron and total ferritin concentrations in renal tissue homogenates were measured on Cobas c311 and e411 analyzers (Roche Diagnostics International AG; Risch-Rotkreuz, Switzerland) according to the manufacturer’s protocols.

Tissue Localization of Iron Ions

Modified Perl’s and Lillie’s methods were applied as previously described^36,37 to localize ferric and ferrous ions in renal tissues, respectively. The protocols involve extra steps to intensify the Prussian blue and Turnbull blue stains by incubating the sections with 3,3′-Diaminobenzidine (DAB) substrate to produce a permanent brown stain.^36,37 All slides were treated with 3% (vol/vol) H₂O₂ in methanol for 60 min before staining with Prussian blue (Santa-Cruz Biotechnology Inc.) and Turnbull blue (Sigma-Aldrich Co.) to control the activities of endogenous peroxidases and catalases on the substrate. Sections were then incubated with DAB (Vector Laboratories, Inc.; Burlingame, CA) for 30 min and later counterstained with hematoxylin.

All sections were examined with EVOS XL Core microscopy (Thermo Fisher Scientific), and a positive reaction was indicated by the presence of a brown color. Quantification of ferric and ferrous ions in all groups was calculated on ImageJ software (https://imagej.nih.gov/ij/) using captured digital images with 100× objective from 15 random non-overlapping microscopic fields from each section (Supplementary Fig. 1).

Immunofluorescence Staining

The co-localization of both ferritin chains in addition to hepcidin with FPN in renal tissues was done by using primary goat (hepcidin and ferritin-H; Santa-Cruz Biotechnology Inc.) and rabbit (ferroportin and ferritin-L; Abcam; Cambridge, MA) polyclonal IgG antibodies. Briefly, all tissue sections were blocked with normal donkey serum for 30 min and were then incubated simultaneously with a mixture of the corresponding primary goat and rabbit antibodies at 1:200 concentration for 3 hr. The slides were subsequently incubated for 60 min with a mixture of tagged cross-adsorbed secondary donkey anti-goat (Alexa Fluor 555) and anti-rabbit (Alexa Fluor 488) IgG antibodies (Thermo Fisher Scientific). All sections were finally counterstained with ProLong Diamond Anti-fade Mountant containing 4′,6-diamidino-2-phenylindole (DAPI; Thermo Fisher Scientific).

The positive control slides consisted of liver sections obtained from normal rats. The negative control slides contained renal sections from the study groups that were treated identically to all other slides, but the primary antibodies were replaced with a mixture of normal rabbit and goat IgG antibodies (Santa-Cruz Biotechnology Inc.). All slides were examined with EVOS FL microscopy (Thermo Fisher Scientific), and digital images were captured within the same session from 10 random non-overlapping fields from each slide using 100× objective. The expression profiles of the target proteins were measured by digital image analysis according to the previously described method³⁸ (Supplementary Fig. 2).

IHC

The localization of TfR1, caspases-3 and -9 in renal tissues, was conducted using primary polyclonal rabbit IgG antibodies (AntibodyPlus Inc.; Brookline, MA) and Elite Vectastain Rabbit ABC kit (Vector Laboratories Inc.) according to the manufacturer’s protocols. The concentrations were 1:100 for TfR1 and 1:150 for caspases. The negative control slides were treated identically to all other slides, but the primary antibodies were replaced with normal rabbit IgG antibodies. The sections were observed with an EVOS XL Core microscope, and images were captured during the same session from 10 random non-overlapping fields for each protein of interest using 100× objective for TfR1 and 40× objective for caspases. The images were then processed for quantification on ImageJ as previously described.^39,40

TUNEL Assay

Cell DNA damage and cell apoptosis/necrosis were assessed in the collected tissue specimens using the Click-iT terminal deoxynucleotidyl transferase-dUTP nick end labeling (TUNEL) Alexa Fluor 488 Imaging Assay (Thermo Fisher Scientific) and by following the manufacturer’s protocol. Stained slides were examined with EVOS FL microscopy using 40× objective. DNA damage was indicated by the emission of green fluorescence dye, and the apoptotic/necrotic cells were counted using the cell counter tool provided with the microscope software (Supplementary Fig. 3). Apoptosis/necrosis index was measured by calculating the percentage of damaged cells in 15 random non-overlapping fields from each tissue section as previously described.⁴¹

ELISA

ELISA was used to measure the concentrations of hepcidin in serum and renal tissue homogenates using a specific rat kit (Cloud-Clone Corp.; Houston, TX). Oxidative stress was assessed by measuring malondialdehyde (MDA; Abcam), total antioxidant capacity (TAC; Abcam), and glutathione peroxidase 4 (GPX4; CUSABIO; Wuhan, China) in renal tissue lysates. Samples were processed in duplicate on an automated ELISA system (Human Diagnostics; Wiesbaden, Germany) and by following the manufacturers’ guidelines for each kit.

Quantitative Reverse Transcription (RT)-PCR

The cDNA was synthesized by transcribing 200 ng of total RNA using a high capacity RNA-to-cDNA RT Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. PCR reactions were carried out in triplicate wells using ABI 7500 system and power SYBR Green master mix (Thermo Fisher Scientific). Each well included 10 µl SYBR Green, 7 µl DNase/RNase free water, 1 µl containing 5 pmol of each primer (Supplementary Table 1), 1 µl cDNA (25 ng), and 40 amplification cycles (95C/15 sec and 60C/1 min) were performed. A minus-RT control from the RT step and another minus-template PCR, in which nuclease-free water was added as a template, were used as negative controls. The 2^−∆∆Ct method was used to perform relative quantitative gene expression of rat hepcidin, ferroportin, TfR1, megalin, cubilin, ferritin heavy and light chains, and caspases-3, -8, and -9 against rat β-actin gene.

Statistical Analysis

SPSS v16 was used for statistical analysis. All data were assessed for normality by the Kolmogorov and Smirnov’s test, and homogeneity by the Levene test. Based on variance equality, one-way ANOVA followed by either Tukey’s honest significant difference (HSD) or Games-Howell post hoc tests were used to compare between the study groups. Correlations were determined by Pearson’s test; p<0.05 was considered significant.

Results

Iron Concentrations in Serum and Renal Tissue Homogenates

No mortality occurred during the study, however, signs of intoxication (e.g., drowsiness, lethargy) were observed from 2 hr and thereafter of acute toxicity only. Serum transferrin saturation (TfSat) and iron levels in addition to renal tissue iron concentrations increased significantly in the chronic group compared with control (p<0.001).

Furthermore, acute intoxication resulted in significantly higher TfSat together with marked elevations in serum and renal tissue iron concentrations compared with the control and chronic groups, and both TfSat (Fig. 1A) and serum iron (Fig. 1B) were maximal in the A-1H group. Renal iron tissue concentrations, however, showed progressive significant increases that reached the highest significant level in the A-8H group (Fig. 1B; p<0.001). Strong positive correlations were also detected between serum and renal iron levels (Fig. 1C). Renal function parameters were marginally but significantly higher in the acute and chronic groups compared with controls (Supplementary Table 2).

Figure 1.

Mean ± SD of (A) serum TfSat, (B) serum and renal tissue iron concentrations in addition to (C) the correlation between serum and renal tissue iron concentrations (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p<0.05 compared with A-8H group). Abbreviation: TfSat, serum transferrin saturation.

Localization of Fe³⁺ and Fe²⁺ Ions in Renal Tissue

While Fe³⁺ iron stain showed low cytoplasmic distribution and basolateral localization in the proximal and distal tubules of controls (Fig. 2A), the distribution of Fe²⁺ was negligible (Fig. 2G). The intensity of Fe³⁺ and Fe²⁺ stains escalated significantly in the renal tubular cells at A-1H (p<0.001) compared with the controls, and both ionic forms were observed as apical and basal cytosolic granules, respectively (Supplementary Fig. 1). The utmost significant increase in the distribution of Fe³⁺ (Fig. 2D) and Fe²⁺ (Fig. 2J) was seen in the A-8H group, and the tubular cells showed many cytoplasmic vacuolations. Furthermore, Fe²⁺ also exhibited nuclear localization in the A-4H (Fig. 2I) and A-8H groups.

Figure 2.

Tissue distribution of Fe³⁺ and Fe²⁺ iron stains in the renal tissues of control (A and G), A-2H (B and H), A-4H (C and I), A-8H (D and J), and chronic (E and K) groups. In addition, (F) Fe³⁺ and (L) Fe²⁺ distribution indexes are shown as bar graphs for the study groups (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p < 0.05 compared with A-8H group). Objective = 100×; scale bars = 5 µm. Abbreviations: Red arrow, cytoplasmic vacuoles; red arrow head, nuclear stain; LH, loop of Henle; PT, proximal tubule; CD, collecting duct; DT = distal tubule.

In the chronic group, both Fe³⁺ (Fig. 2E) and Fe²⁺ (Fig. 2K) ions were also significantly higher than controls (p<0.001), and the cellular distribution of Fe³⁺ was also significantly larger compared with all acute groups (Fig. 2F; p<0.001). In contrast, Fe²⁺ accumulated in cytoplasmic vacuoles, showed no nuclear staining, and was significantly lower compared with the A-8H group (Fig. 2K; p<0.001).

The Expression of Iron-homeostatic Proteins in Renal Tissue

TfR1, Cubilin and Megalin Receptors

The primary antibodies against TfR1 clearly labeled the apical and subapical regions, with occasional faint basal expression, of renal tubular epithelium obtained from the controls (Fig. 3A). The immunostain of TfR1 showed a biphasic expression pattern characterized by an initial significant decrease until the A-2H group (Fig. 3B), followed by a marked stronger immunostain in the remaining groups of acute toxicity that, however, showed comparable levels with controls (Fig. 3F). The TfR1 immunostain was also significantly weaker in the chronic group (Fig. 3E) than the controls (p<0.001).

Figure 3.

IHC localization of TfR1 in the renal tissues of control (A), A-2H (B), A-4H (C), A-8H (D), and chronic (E) groups. In addition, mean ± SD of (F) the IHC arbitrary scores of TfR1 expression together with the relative mRNA expression of (G) TfR1, (H) cubilin, and (I) megalin receptors (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p<0.05 compared with A-8H group). Objective = 100×; scale bars = 2 µm. Abbreviations: TfSat, serum transferrin saturation; DT, distal tubule; PT, proximal tubule; CD, collecting duct.

In parallel, TfR1 mRNA declined significantly (p<0.001) in the early hours of acute toxicity followed by a marked upsurge that reached the highest level in the A-8H group (Fig. 3G). In addition, chronic toxicity resulted in a significant decline in the mRNA of TfR1 compared with controls (Fig. 3G; p<0.001). Interestingly, significant upregulations in the mRNAs of both cubilin (Fig. 3H) and megalin (Fig. 3I) receptors were observed in the acute subgroups compared with control, and the maximal significant value for both receptors was seen in the A-4H group (p<0.001). The mRNA of both receptors also increased significantly following chronic intoxication compared with the control group (p<0.01).

Serum and Renal Tissue Ferritin

Serum ferritin increased significantly following acute and chronic toxicity, and the latter group had the highest level compared with controls (p<0.001). In contrast, the concentrations of total renal ferritin showed significant declines in all acute subgroups except for the A-1H group (Fig. 4; panel 1). Significant correlations were also observed between serum and renal tissue ferritin concentrations that were paradoxically negative and positive with acute (Fig. 4, panel 4) and chronic intoxication (Fig. 4, panel 5), respectively.

Figure 4.

Mean ± SD of (1) serum and renal tissue concentrations of ferritin, (2) immunofluorescence arbitrary scores of ferritin H- and L-chains, and (3) relative mRNA expression of ferritin H- and L-chains in renal tissues and correlations between serum and renal tissue ferritin concentrations in (4) acute and (5) chronic iron toxicity (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p<0.05 compared with A-8H group). In addition, panels A to T show the immunofluorescence localization of the H- (red) and L- (green) chains in the renal tissues of control (A–D), A-2H (E–H), A-4H (I–L), A-8H (M–P), and chronic (Q–T) groups. All sections were counterstained with DAPI. Objective = 100×; scale bars = 2 µm. Abbreviations: DT, distal tubule; PT, proximal tubule; LH, loop of Henle; CD, collecting duct; DAPI, 4′,6-diamidino-2-phenylindole.

The immunofluorescence results showed that the ferritin H- (Fig. 4B) and L- (Fig. 4C) chains were homogeneously localized in the cytoplasm of renal tubular cells of control tissues, and the H-chain appeared dominant (Fig. 4D). The early significant changes were detected in the A-2H group compared with controls that were characterized by a paradoxical decrease in the H-chain (Fig. 4F; p<0.001) and an increase in the L-chain (Fig. 4G; p<0.001), and the latter was more dominant (Fig. 4H). The expression of both chains then declined significantly until the 8th hr of acute toxicity (Fig. 4; panel 2), and, remarkably, the L-chain appeared aggregated in intracytoplasmic vesicles (Fig. 4K). In contrast, the highest expression levels for the H- (Fig. 4R) and L-chains (Fig. 4S) were observed in the chronic group (Fig. 4; panel 2). The mRNA expressions of both chains correlated with the immunofluorescence results and showed a similar fluctuation pattern between the different acute subgroups, and the highest expression for both chains was observed in the chronic group (Fig. 4; panel 3).

Serum and Renal Hepcidin and the Expression of FPN by Renal Tissue

Serum and renal hepcidin increased steadily and progressively following acute overloading, reaching their peaks at 8 hr (Fig. 5; panel 1). Both serum and renal hepcidin were also significantly higher following chronic intoxication than controls (Fig. 5; panel 1). Renal hepcidin concentrations also correlated strongly and positively with serum hepcidin (Fig. 5; panel 4), serum iron (Fig. 5; panel 5), and renal tissue iron (Fig. 5; panel 6).

Figure 5.

Mean ± SD of (1) serum and renal tissue concentrations of hepcidin, (2) immunofluorescence arbitrary scores of hepcidin and FPN, and (3) relative mRNA expression of hepcidin and FPN in renal tissue together with the correlations between renal tissue hepcidin concentrations with (4) serum hepcidin, (5) serum iron concentrations, and (6) renal tissue iron concentrations (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p<0.05 compared with A-8H group). In addition, panels A to T show the immunofluorescence localization of hepcidin (red) and FPN (green) proteins in the renal tissues of control (A–D), A-2H (E–H), A-4H (I–L), A-8H (M–P), and chronic (Q–T) groups. All sections were counterstained with DAPI. Objective = 100×; scale bars = 2 µm. Abbreviations: DT, distal tubule; CD, collecting duct; PT, proximal tubule; FPN, ferroportin; DAPI, 4′,6-diamidino-2-phenylindole.

Hepcidin (Fig. 5B) and FPN (Fig. 5C) were mainly localized basolaterally by immunofluorescence in renal tubular cells of control tissues, and hepcidin showed vesicular constellations. The initial significant alteration was detected in the A-2H group where the expression of hepcidin increased (Fig. 5F) while FPN decreased (Fig. 5G) significantly, and both were located at the basal membrane of tubular epithelium (Fig. 5H). In addition, hepcidin (Fig. 5J) reached its highest expression in the A-4H group that also showed a further decline in FPN (Fig. 5K). Hepcidin was clearly localized in the basal circumference of tubular cells with few patches of cytoplasmic expression that were also occasionally seen in the apical border (Fig. 5L). However, FPN showed weak circumferential basal localization and limited cytoplasmic expression. In the A-8H group, a significant decrease in hepcidin (Fig. 5N) coincided with a significant increase in FPN (Fig. 5O). The chronic group also had significantly higher hepcidin (Fig. 5R) and FPN (Fig. 5S) expressions compared with the control group (Fig. 5; panel 2).

The mRNA expression of hepcidin reached its highest significant level (7 folds) in the A-2H group followed by persistent significant declines until the A-8H (Fig. 5; panel 3). In the opposite direction, FPN mRNA showed the lowest expression level (9 folds) in the A-2H group followed by steady significant escalations that reached the maximum in the A-8H group (Fig. 5; panel 3). The mRNA of both molecules was also significantly increased in the chronic group compared with controls (Fig. 5; panel 3).

Renal Tissue Concentrations of Oxidative Stress Markers

The oxidative stress markers in renal tissue homogenates from the early groups of acute toxicity showed comparable levels with controls (Supplementary Table 1). In the A-4H group, MDA levels increased significantly, while TAC and GPX4 declined significantly, compared with the control group (p<0.001 for all markers). The highest concentrations of MDA were seen in the A-8H group and were associated with the lowest levels of TAC and GPX4. The chronic group also showed a significant increase in MDA that concurred with significant reductions in TAC and GPX4 compared with all groups, except the A-8H group (Supplementary Table 1).

Iron-induced Renal Cell Damage and the Expression of Caspases-3 and -s9

A minority of renal cells were positive for apoptosis in the control group, and the earliest significant cell damage was detected in the A-4H group (Fig. 6D), which reached its highest numbers in the A-8H group (Fig. 6G). Similarly, samples from chronic overloading (Fig. 6J) showed higher numbers of apoptotic cells compared with control (Fig. 6; panel 1). Significant correlations for the oxidative stress markers and apoptosis index in renal tissues were detected with serum and kidney iron, and hepcidin concentrations (Table 2).

Figure 6.

Apoptotic bodies by TUNEL assay (left column) together with the immunohistochemical expression of caspases-9 (middle column) and -3 (right column) in the renal tissues of control (A–C), A-4H (D–F), A-8H (G–I), and chronic (J–L) groups. In addition, mean ± SD of (1) apoptosis index, IHC arbitrary scores of (2) caspase-9 and (3) caspase-3, together with relative mRNA expression of (4) capsase-8, (5) caspase-9, and (6) caspase-3 (a = p<0.05 compared with control, b = p<0.05 compared with A-1H group, c = p<0.05 compared with A-2H group, d = p<0.05 compared with A-4H group, and e = p<0.05 compared with A-8H group). Objective = 40×; scale bars = 10 μm (left column) and 15 µm (middle and right columns). Abbreviations: TUNEL, terminal deoxynucleotidyl transferase-dUTP nick end labeling; red star, glomerulus; red arrow head, renal tubule.

In agreement with the TUNEL’s results, the immunoexpression of caspases-9 (Fig. 6B) and -3 (Fig. 6C) was low in control tissue. Significantly greater immunostain was detected for caspases-9 and -3 in the A-2H (Fig. 6; panels 2 and 3), and the highest expression profiles for both enzymes were seen in the A-8H group (Fig. 6H and I). In addition, the chronic group had significantly stronger immunostains for Casp-9 (Fig. 6K) and -3 (Fig. 6L) compared with controls but was weaker than the A-8H group (Fig. 6; panels 2 and 3). The mRNAs of caspases-8 (Fig. 6; panel 4), -9 (Fig. 6; panel 5), and -3 (Fig. 6; panel 6) were significantly upregulated post-acute intoxication, and the highest levels for all genes were observed in the A-8H group. The mRNAs of the 3 enzymes were also significantly increased in the chronic group compared with controls but were significantly lower than the A-8H group.

Discussion

Herein, we measured the effects of acute sublethal and chronic iron intoxications on the expression profile of cellular iron-homeostatic molecules together with iron accumulation and ferroptosis in normal kidney. The results showed significant escalations in both serum and renal tissue iron concentrations following acute and chronic overloading, but the plasma levels in both models did not exceed the saturation capacity of Tf. Coherently, the intracellular uptake of iron by the renal tissues was documented as early as 1-hr post-acute ingestion, and there was a significant strong positive correlation between plasma and renal iron levels. Our results provide further support to the notion that the kidney is dynamically involved in systemic iron homeostasis^{12,16,18,32,42} and suggest that renal tubules favor iron reabsorption despite excess plasma concentrations.

Renal tubules, under physiological conditions, reabsorb 99.3% of filtrated TBI/day^17,19 through TfR1 and cubilin/megalin complex, which modulate the uptake of Tf by endocytosis.^21,43,44 However, the affinity of TfR1 to its ligand is 100 times greater than the cubilin/megalin complex.^21,45 Interestingly, an in vitro study has reported that Tf was preferentially endocytosed by the cubilin/megalin complex during excess iron supplementation, while TfR1 was dominant under restricted conditions.²² Others have also shown that both cubilin and megalin receptors increased significantly in a mouse model of hemolytic anemia, suggesting their contribution to scavenging surplus filtrated iron.³²

Our study, similarly, demonstrated that both toxicity models induced marked decreases in renal TfR1 gene and protein alongside upregulations in the mRNAs of cubilin/megalin receptors. These observations correlate with the earlier reports and withstand the proposition that the renal mechanisms for handling iron overloading involve increasing transcytosis of excess filtrated TBI by tubular cells through the cubilin/megalin receptors rather than TfR1.^22,32,46 A possible explanation could be related to the differential regulatory effects of iron on the expression of the designated receptors.^21,46,47 However, more studies are needed to explore the roles of Tf-endocytotic pathways on systemic iron homeostasis, because TfR1 leads to Tf recycling while the cubilin/megalin pathway induces lysosomal degradation of the protein.^48,49

Following endocytosis, TBI is processed by intracellular lysosomes to extract iron from Tf in the form of Fe²⁺.^1,7,44 Cytosolic Fe²⁺, also known as the intracellular iron labile pool, is redox-active and could induce cell injury by catalyzing the Fenton’s reaction and, thus, producing ROS, oxidative stress, and ferroptosis.^13–15 Cells have, therefore, evolved several regulatory mechanisms to neutralize the toxic effects of intracellular Fe²⁺.^17,46 Ferritin, the main intracellular iron-storage protein, consists of H- and L-chains, and Fe²⁺ is incorporated within the H-chain in the nontoxic form, Fe³⁺, following its nucleation by the L-chain.^9–12 However, iron overloading triggers the degradation of intracellular ferritin through ferritinophagy.^13,50 In addition, the levels of serum ferritin, which is mainly composed of L-chains, increase markedly in response to iron overloading.^51–53

Cohen et al. have reported that renal PT and macrophages are the main sources of circulatory ferritin,⁹ and marked increases in both serum ferritin and the renal expression of the L-chain were also observed in mice following the deletion of the H-chain gene in the PT.⁵⁴ AKI similarly resulted in a significant decrease in the H-chain of hepcidin knockout mice.³¹ In contrast, Starzynski et al. have reported an increase in the renal expression of ferritin peptides, and the H-chain was fully saturated with non-heme iron in mice lacking the HO-1 gene.²⁴ Concurrently, gene knockdown of hepcidin, hemojuvelin, and the simultaneous deletion of both hephaestin with ceruloplasmin genes in mice, also led to significant increases in serum and renal ferritin together with pronounced iron depositions in the kidney.^42,55

The current study demonstrated paradoxical results for renal total ferritin as acute overloading induced marked declines while the chronic group resulted in a significant elevation. However, serum ferritin increased significantly in both models and correlated inversely and positively with renal ferritin in the acute and chronic models, respectively. We, therefore, speculate that acute iron overloading may either induce disassembly of both ferritin chains, and/or paradoxically upregulate the L- and inhibit the H-chain expressions in renal tubular cells followed by the release of L-chain into systemic circulation, thus, resulting in the observed inverse relationship between serum and renal tissue ferritin.^9,13,51,52 In contrast, chronic iron overload may offer the necessary time for renal cells to compensate for the excessive intracellular accumulations of reabsorbed iron together with the release of the L-chain into circulation by increasing the production of both ferritin chains.^42,55

Hepcidin is the main iron regulatory hormone through its receptor, FPN, and both molecules are expressed by renal tubular cells.^5,23,27 Experimental studies on iron dyshomeostasis have shown significant simultaneous increases in renal hepcidin and FPN that coincided with the deposition of iron in renal tissues.^24,31,42 Opposingly, Xie et al. have also reported inhibition of renal tissue FPN, at the mRNA and protein levels, following acute renal injury that concurred with an increase in renal iron deposition.³³ Furthermore, others have shown no effect on FPN mRNA and/or protein in renal tissues following either the dual deletion of hephaestin and ceruloplasmin genes in mice,⁵⁵ using iron-restricted diet in rats,²⁰ or excess iron conditions in vitro.¹⁸

The levels of serum and renal hepcidin in the present study were significantly elevated in both acute and chronic groups, and renal hepcidin correlated more significantly with renal iron concentrations. Furthermore, the expression of renal FPN was paradoxically down- and upregulated following acute and chronic intoxication, respectively. Therefore, we hypothesize that the expression of renal hepcidin might be under the control of the levels of iron trafficking within renal tubules rather than circulatory iron. In addition, we suggest that acute intoxication might induce the degradation of FPN, the only known iron exporter protein, subsequently resulting in iron entrapment within renal tubular cells.^31,33 On the contrary, the increase of FPN in tubular epithelium during chronic toxicity proposes that iron exportation to the circulation is still, probably, operative and could consequently diminish the levels of cytosolic Fe²⁺.^2,55

The present findings also revealed that the levels of intracellular Fe²⁺ were greater with acute whereas Fe³⁺ was significantly higher in chronic overloading, suggesting that the forms of accumulated iron ions within renal cells could be dependent on the model of intoxication. Currently, the deposition of Fe²⁺ in renal tissues was only measured by a single study, in which the researchers adopted a similar histological technique to the current study, and they have shown significant increases in Fe²⁺ following renal ischemia-reperfusion.³⁷ Remarkably, the remaining earlier studies that investigated the intercommunications between renal pathologies and iron dysregulation only targeted Fe³⁺ as a marker of renal injuries.^24–26,42 Although Fe³⁺ depositions could indirectly indicate a simultaneous increase in intracellular Fe²⁺, we believe that the direct detection of Fe²⁺ iron is more accurate to reflect on iron-induced renal pathologies,³⁷ especially that the histological staining protocols are identical.^36,37

Several studies have reported that iron plays a major role in kidney injury, and the metal has been a target for therapeutic intervention.²⁷ Excess cytosolic Fe²⁺ may cause cell death by caspase-dependent (mitochondrial-mediated) and/or independent (ferroptosis) pathways.^13,14 Iron-induced mitochondrial damage could be initiated either by intrinsic (caspase-9) or extrinsic (caspase-8) pathways that finally activate caspase-3, which is central for cell apoptosis.⁵⁶ In agreement with these reports, our results showed marked elevations in the markers of oxidative stress and renal cell apoptosis that were more pronounced in the acute than chronic intoxication. In addition, there was a marked decline in the levels of renal GPX4, a well-established marker of ferroptosis.⁵⁷

Our observations advocate that the more prominent cellular damage following acute intoxication could be related to the increase in cytosolic Fe²⁺, possibly following the inhibition of both FPN and ferritin, thus, leading to marked oxidative stress and ferroptosis.^13,14 On the contrary, the higher levels of ferritin during chronic overloading could provide cytoprotection, presumably by sequestrating excess iron in the Fe³⁺ state.¹² Furthermore, excess cytosolic Fe²⁺ in chronic toxicity could also be compensated through the observed increase in the exporter protein, FPN, and thus reducing its toxic effects.^6,18 Therefore, the differences in ferroptosis/apoptosis rates between acute and chronic intoxications could be dependent upon the balance between renal iron reabsorption against renal ferritin storage and/or FPN exportation capacities.

In conclusion, this study provides further support to the significance of the kidney in systemic iron homeostasis and advocates that renal tubules adopt reuptake of filtrated TBI, despite the increase in plasma levels, mainly through cubilin/megalin receptors. In addition, renal FPN and ferritin are differentially regulated by acute and chronic iron toxicity and their inhibition following acute toxicity could be associated with elevations in cytosolic Fe²⁺, oxidative stress, and ferroptosis. Opposingly, the higher levels of FPN and ferritin in chronic toxicity might modulate the tissue damage induced by excess cytosolic Fe²⁺ through enhancing its exportation and/or incorporation into the ferritin shell. However, ferroptosis could still occur during chronic overloading if the rate of renal iron reabsorption exceeded renal ferritin storage and/or FPN exportation capacities. More studies are still needed to illustrate the interconnections between the kidney and systemic iron dyshomeostasis.