Toll-Like Receptor 4/MyD88–Mediated Signaling of Hepcidin Expression Causing Brain Iron Accumulation, Oxidative Injury, and Cognitive Impairment After Intracerebral HemorrhageClinical Perspective
Background: Disturbance of brain iron metabolism after intracerebral hemorrhage (ICH) results in oxidative brain injury and cognition impairment. Hepcidin plays an important role in regulating iron metabolism, and we have reported that serum hepcidin is positively correlated with poor outcomes in patients with ICH. However, the roles of hepcidin in brain iron metabolism after ICH remain largely unknown.
Methods: Parabiosis and ICH models combined with in vivo and in vitro experiments were used to investigate the roles of hepcidin in brain iron metabolism after ICH.
Results: Increased hepcidin-25 was found in serum and primarily in astrocytes after ICH. The brain iron efflux, oxidative brain injury, and cognition impairment were improved in Hepc−/− ICH mice but aggravated by the human hepcidin-25 peptide in C57BL/6 ICH mice. Data obtained in in vitro studies showed that increased hepcidin inhibited the intracellular iron efflux of brain microvascular endothelial cells but was rescued by a hepcidin antagonist, fursultiamine. Using parabiosis ICH models also shows that increased serum hepcidin prevents brain iron efflux. In addition, Toll-like receptor 4 (TLR4)/MyD88 signaling pathway increased hepcidin expression by promoting interleukin-6 expression and signal transducer and activator of transcription 3 phosphorylation. TLR4−/− and MyD88−/− mice exhibited improvement in brain iron efflux at 7, 14, and 28 days after ICH, and the TLR4 antagonist (6R)-6-[N-(2-chloro-4-fluorophenyl) sulfamoyl] cyclohex-1-ene-1–carboxylate significantly decreased brain iron levels at days 14 and 28 after ICH and improved cognition impairment at day 28.
Conclusions: The results presented here show that increased hepcidin expression caused by inflammation prevents brain iron efflux via inhibition of the intracellular iron efflux of brain microvascular endothelial cells entering into circulation and aggravating oxidative brain injury and cognition impairment, which identifies a mechanistic target for muting inflammation to promote brain iron efflux and to attenuate oxidative brain injury after ICH.
The secondary brain injury caused by accumulative brain iron after intracerebral hemorrhage (ICH) has become well recognized.1 Iron, as one of the most important degradation products of a hematoma, not only can cause direct toxic injury2,3 and damage the DNA4,5 in the acute phase of ICH but also can lead to cognition impairment in the later phase of ICH.6–8 Accordingly, promoting accumulative brain iron efflux is beneficial to ICH treatment. This has been demonstrated by the application of an iron chelator such as deferoxamine in multiple animal ICH models.9–11 One experiment showed that deferoxamine reduced brain iron overload after ICH without improving neurological deficits in rats.12 Although no serious adverse events occurred in patients with acute ICH13 and although deferoxamine is currently in clinical trials,14 the effects on patients with ICH remain unknown. The unknown mechanisms of brain iron metabolism greatly limit the development of therapeutic drugs targeting brain iron efflux and its therapeutic effects after ICH. Therefore, elucidating the mechanisms underlying brain iron metabolism can give us more promising therapeutic targets for ICH treatment on brain iron efflux. Moreover, the promotion of sharply accumulated brain iron entering into circulation via the blood-brain barrier can solve the problem of brain damage caused by iron after ICH.
Hepcidin, an important iron-regulative peptide hormone,15 plays a crucial role in regulating iron metabolism by controlling intracellular iron efflux via binding and internalizing ferroportin (the only known efflux channel of cells).16 Among the hepcidin-20, -22, and -25 peptides, only hepcidin-25 has a biofunction.16 Previously, on the basis of clinical data from patients with ICH, we concluded that serum hepcidin levels are positively correlated with poor outcomes and that peak levels of serum hepcidin were observed in patients at 3 days after ICH onset,17 which seems to correlate with the changing trends of inflammation after ICH. These results suggest that hepcidin may be involved in brain iron metabolism after ICH. However, the exact roles of hepcidin in brain iron metabolism after ICH still need to be explored.
In general, hepcidin can be upregulated by inflammatory responses16 such as the interleukin-6 (IL-6)/janus kinase/signal transducer and activator of transcription 3 (STAT3) signaling pathway.18,19 Previously, we and others reported that Toll-like receptor 4 (TLR4)–mediated inflammation is one of the most important contributors to brain injury caused by ICH20–22 and that IL-6 is one of the downstream inflammatory cytokines of TLR4 signaling.21 Therefore, inflammation caused by ICH may influence the upregulated expression of hepcidin. In addition, minocycline, as a microglia activation inhibitor,23 is shown to markedly reduce accumulative iron after ICH.24 These results strongly suggested that the changes in hepcidin caused by inflammation may be involved in brain iron metabolism after ICH. Accordingly, shedding light on the mechanisms underlying hepcidin changes and those involving brain iron metabolism can achieve better therapeutic effects on brain iron evacuation and the damaging effects of ICH.
In this study, we are the first to show that the TLR4/MyD88 signaling pathway increased both brain astrocyte-derived and serum hepcidin, which resulted in the inhibition of brain iron efflux into the circulation after ICH via binding to the ferroportin of brain microvascular endothelial cells (BMVECs) that reduce iron efflux channels. These results indicate that a therapeutic target for inflammatory responses may be beneficial not only to reduce inflammatory injury but also to alleviate the accumulated brain iron caused by oxidative injury after ICH.
A total of 1672 mice were used in our study. C57BL/6 male mice (weight, 20–24 g) were purchased from the Animal Center of Daping Hospital, Third Military University (Chongqing, China). Hepc+/ − mice were purchased from the Mutant Mouse Resource Research Centers (Davis, CA). We interbred Hepc heterozygous mice to produce and identify knockout and wild-type offspring according to the following gene primer: DNA292-9 5’–TTGCACGGGGAAGAAAGCAG, DNA292-10 5’–GACCTGTAAACCCAGCTCAG, Neo3a 5’–GCAGCGCATCGCCTTCTATC. TLR4−/− and MyD88−/− mice (weight, 20–24 g) were purchased from the Model Animal Research Center of Nanjing University (Nanjing, Jiangsu). TRIF−/− mice (weight, 20–24 g) were purchased from Jackson Laboratories (Bar Harbor, ME). All mice had a C57BL/6 genetic background. The mice were housed in a specific pathogen-free environment with 12-hour light/dark cycles, a controlled temperature (25°C), and relative humidity (45%–55%) with free access to standard food and water. To exclude the influence of sex on brain iron metabolism, male mice were used in this study. The physiological parameters of these mice have no significant difference (online-only Data Supplement Table I). The Animal Ethics Committee of Third Military Medical University approved all animal experients. All experiments were performed according to national guidelines and Animal Research: Reporting of In Vivo Experiments guidelines.25
According to the previous report,26 male mice of the same weight and size were paired ≥2 to 3 weeks before the surgery to ensure that they were compatible. In the present study, we successfully constructed the C57BL/6-C57BL/6 isochronic parabiosis models, C57BL/6-Hepc−/− heterochronic parabiosis models, and Hepc−/−-Hepc−/− isochronic parabiosis models, which we used to investigate the role of periphery hepcidin in brain iron metabolism when subjected to ICH.
The detailed procedures used to construct the ICH model were established in our previous studies.27,28 The parabiosis mice ICH models were constructed ≈20 days after surgery because their blood chimerism occurs 10 to 14 days after the surgery.26 Mice were briefly anaesthetized intraperitoneally with 4% phenobarbital sodium and were immobilized on the stereotaxic apparatus (RWD Life Science Co, Shenzhen, China). All experimental mice, including each of the 2 mice from the parabiosis models, received a total of 20 μL autologous blood injected into the caudate nucleus (bregma 0: 0.8 mm anterior, 2 mm left lateral, and 3.5 mm deep) successively. Unsuccessful ICH models, including asymptomatic and dead mice before euthanasia, were excluded from this study. Before the ICH mice were euthanized, we collected all of their blood through removal of an eyeball. The blood was untreated with anticoagulants to avoid influencing serum iron levels. The blood samples were placed at 4°C for 1 hour and centrifuged after clotting to obtain the serum samples, and then they were stored in aliquots at −80°C.
Human Hepcidin-25 Peptide and ETHYL (6R)-6-[N-(2-Chloro-4-Fluorophenyl) Sulfamoyl] Cyclohex-1-ene-1–Carboxylate Application
We intravenously administered human hepcidin-25 peptide (Abcam) at a dose of 1 μg/dL once daily for 5 successive days beginning 6 hours after ICH. According to our previous report,28 (6R)-6-[N-(2-chloro-4-fluorophenyl) sulfamoyl] cyclohex-1-ene-1–carboxylate (TAK242; Takeda Pharmaceutical Co) was formulated with 1% dimethyl sulfoxide (σ) and double-distilled water to a final concentration of 0.4 mg/mL and then injected intraperitoneally at a dose of 3 mg/kg once daily for 5 successive days beginning 6 hours after ICH.
Brain Nonheme Iron Measurement
Using the protocol reported by Auriat and colleagues12 with modifications, we measured the brain nonheme iron after ICH. Experimental mice were briefly overdosed with phenobarbital sodium and perfused with PBS. The brains were removed and dissected into ipsilateral brain tissues and contralateral similar tissues. To avoid metal contamination, Teflon-coated blades (7280 L, Thermo Scientific, Kalamazoo, MI) and plastic forceps were used to handle the brain samples. After the samples were weighed, a ratio of 1:10 wt/vol distilled water was added, and then the samples were homogenized in a Dounce homogenizer. Two aliquots of the solution were then transferred to separate microcentrifuge tubes. An equal volume of protein precipitate solution (1N HCl, 10% trichloroacetic acid in dH2O) was added to each aliquot, which was mixed and heated for 1 hour at 95°C in an sealed tubes followed by an 8200g, 10-minute centrifuge. Then, 100 μL supernatant was transferred into 2 separate new microcuvettes, and 600 μL chromogen solution (0.025 mmol/L ferrozine, 3 mol/L sodium acetate, 1.5% [vol/vol] thioglycolic acid in dH2O) was added and mixed. After being left for 30 minutes at room temperature to allow color development, the samples were measured at 562 nm with an automatic microplate reader (Thermo Scientific Varioskan Flash).
Diaminobenzidine-Enhanced Turnbull Staining With Neural Cells
The modified diaminobenzidine-enhanced Turnbull staining protocol29,30 was applied for detection of brain iron and neural cells. After deparaffinization in xylene and rehydration, sections were immersed in aqueous ammonium sulfide solution (2% in deionized water) for 90 minutes for detection of total nonheme iron. This process was followed by a 30-minute incubation with aqueous solution containing 10% potassium ferricyanide and 0.5% HCl. After 3 washing steps with deionized water, sections were incubated in methanol containing 0.01 mol/L sodium azide and 0.3% hydrogen peroxide for 60 minutes for blocking of endogenous peroxidase. Next, sections were washed with 0.1 mol/L phosphate buffer and 0.3% hydrogen peroxide for 60 minutes for blocking of endogenous peroxidase. Iron staining was amplified by a solution containing 0.025% diaminobenzidine (Sigma Aldrich) and 0.005% hydrogen peroxide in 0.1 mol/L phosphate buffer for 20 minutes. The reaction was stopped by rinsing the sections in tap water. After development, sections were washed with deionized water and steamed for 30 minutes in EDTA buffer (pH 8.5). After overnight incubation of rabbit anti-NeuN (1:100; Millpore), rabbit anti–glial fibrillary acidic protein (GFAP) antibodies (1:100; Lispan) or rabbit anti–IBa-1 antibodies (1:100; Lispan) at 4°C. The next day, after being washed 3 times with PBS, sections were incubated with avidin-conjugated alkaline phosphatase (1:500, Sigma Aldrich) and developed with Fast Blue substrate (Sigma) to produce a blue signal for neuron (NeuN), astrocyte (GFAP), or microglia (IBa-1). Sections were dehydrated and mounted on coverslips with Permount mounting medium.
Analyses were performed with SPSS 13.0 software, and all values are presented as mean±SD or percentage. A χ2 test was used to compare the differences in proportions between 2 groups. Student t test and 1-way ANOVA followed by the Scheffé post hoc test were used to compare differences between ≥2 groups. Two-way ANOVA with repeated measures were performed when appropriate to compare repeated measured data, and the main effects of the genotype (treatment) and time points and the interaction were assessed between the two. A value of P<0.05 was considered statistically significant.
An expanded Methods section containing a detailed description of brain, serum hepcidin and serum iron measurement, Western blot, immunohistochemical and immunofluorescent staining and analysis, determination of dichlorofluorescein and 8-isoprostane, brain water content, neurological deficient score, radial arm water maze (RAWM), and cell culture and intracellular ferritin measurement is provided in the online-only Data Supplement.
Upregulated Hepcidin Expression in Both Brain and Serum After ICH
Given that hepcidin plays an important role in regulating iron metabolism15 and that hepcidin-25 has a biofunction,16 we investigated brain and serum hepcidin-25 expression in ICH C57BL/6 mice. First, we detected brain hepcidin expression using a Western blot and found that the hepcidin levels were increased beginning 12 hours after ICH with a peak level at day 3 compared with sham groups and normal C57BL/6 mice (NS; Figure 1A). Similar results were also obtained with hepcidin ELISA kits (Figure 1B). In line with our previous findings of serum hepcidin changes in patients with ICH,16 increased serum hepcidin levels were also found in ICH mouse models with a peak at day 1 after ICH (Figure 1C). In addition, the results of immunohistochemical staining showed that hepcidin-positive cells were significantly increased in brain tissues at day 3 after ICH (Figure 1D). Furthermore, immunofluorescent staining was used to find the cell sources of hepcidin, and the results showed that expression of hepcidin was mostly colocalized with GFAP-positive astrocytes, with little in CD11b-positive microglia (Figure 1E). In contrast, the Hepc−/− mice did not express hepcidin after ICH (online-only Data Supplement Figure I). These results showed that increased brain hepcidin existed primarily in astrocytes after ICH. It is intriguing that we found that the expression of hepcidin encircled CD31-positive BMVECs without colocalization (Figure 1E, red arrow). To further investigate whether the positive staining of hepcidin and GFAP astrocytes adjacent to endothelial cells is peripheral production of hepcidin and uptake by perivascular astrocytes, fluorescence in situ hybridization experiments were performed, and the results showed that the glial cell–like hepcidin mRNA existed in perihematomal tissues at 3 days after ICH (online-only Data Supplement Figure II). These results suggest that increased hepcidin may also play a critical role in brain iron metabolism after ICH.
In addition, to understand the iron-changing features in the brain and during circulation under ICH, we detected the brain iron contents from 12 hours to 28 days after ICH. The results showed that brain iron gradually increased from day 3 after ICH with a peak level at day 14 (online-only Data Supplement Figure IIIA) and remained at a high level until day 28. Moreover, using the methods of diaminobenzidine-enhanced Turnbull staining with neural cells, we found that the accumulated brain iron existed in neurons, microglias, astrocytes, and endothelial cells at day 14 after ICH (online-only Data Supplement Figure IIIB). Then, we investigated the serum iron levels in ICH mice and found that serum iron levels were also decreased from day 1 to 7, especially at day 1 after ICH (online-only Data Supplement Figure IIIC).
Hepc−/− Mice Exhibited Improvement of Brain Iron Efflux, Brain Oxidative Injury, and Cognition Impairment After ICH
Next, we investigated the roles of hepcidin in brain iron metabolism after ICH. First, we found no significant difference in brain iron levels between Hepc−/− and C57BL/6 normal mice (Figure 2A). Then, using the Hepc−/− mice ICH models, we found that the accumulative iron in the brain was decreased compared with C57BL/6 mice (Figure 2B), whereas serum iron contents were increased (Figure 2C). We then investigated the effects of reduced brain iron from 12 hours to 28 days after ICH by determining the production of reactive oxygen species using the index of 2′,7′-dichlorodihydrofluorescein (DCFH)31 and 8-isoprostane.32 Our findings showed that the contents of DCFH and 8-isoprostane in brain were increased after ICH, whereas they were less in Hepc−/− mice than in the C57BL/6 mice (Figure 2D and 2E). We subsequently investigated the brain water contents and found that the brain water contents at days 3, 5, and 7 after ICH were decreased in Hepc−/− mice compared with C57BL/6 mice (Figure 2F), as were the neurological deficient scores between the 2 groups (Figure 2G). Furthermore, we measured the effects of reduced iron contents at 28 days after ICH using the RAWM to confirm iron-related cognition impairment and found that Hepc−/− and C57BL/6 mice showed similar spatial learning for the visible platform during training stage, whereas Hepc−/− mice showed enhanced learning and memory for the hidden platform location compared with the C57BL/6 mice during the testing phase (Figure 2H).
Application of Human Hepcidin-25 Peptide Aggravated Brain Iron Accumulation, Brain Oxidative Injury, and Cognition Impairment After ICH
To discover the roles of hepcidin in brain iron metabolism and its effects after ICH, human hepcidin-25 peptide was used in the C57BL/6 mice once daily for 5 consecutive days beginning 6 hours after ICH. We found that brain iron accumulation was increased in hepcidin-treated groups compared with saline-treated ICH mice (Figure 3A). The reactive oxygen species index of brain DCFH (Figure 3B) and 8-isoprostane levels (Figure 3C) were also higher in the hepcidin-treated mice than in the saline-treated mice after ICH. In addition, we found that the brain water contents at days 3, 5, 7, and 14 after ICH (Figure 3D) and neurological deficient scores at days 3, 5, and 7 after ICH (Figure 3E) were increased in the hepcidin-treated mice compared with the saline-treated mice after ICH. Last, the hepcidin-treated mice showed poorer learning and memory for the hidden platform location than the saline-treated mice during the testing phase in the RAWM test (Figure 3F).
Increased Hepcidin Inhibits BMVECs Intracellular Iron Efflux In Vitro
The pathway across blood-brain barrier from brain to circulation is one of the major pathways for iron efflux after ICH, and increased hepcidin was shown to encircle BMVECs with the use of immunofluorescent staining after ICH (Figure 1E). In addition, hepcidin has been shown to regulate the cellular iron efflux by binding to ferroportin and inducing its internalization.16 It has been shown that ferroportin widely localized in most of neural cells with the immunofluorescent staining such as neurons, microglias, astrocytes, and endothelial cells33,34 (online-only Data Supplement Figure IV). Accordingly, we explored how the increased hepcidin influenced brain iron efflux to the brain in vitro. Astrocytes were cultured and stimulated by 5 μmol/L hemoglobin to mimic in vitro ICH models based on our previous reported methods,35 and BMVECs were stimulated by supernatants from astrocytes for 12 hours after being incubated with ferric ammonium citrate for 24 hours and washed with iron-free DMEM (Figure 4A). In addition, the BMVEC intracellular ferritin content, the cytosolic iron storage protein, was measured at 24 and 36 hours after the beginning of BMVEC culture (Figure 4A). We found that when BMVECs were stimulated by C57BL/6 astrocyte supernatants, the remaining intracellular ferritin levels were higher in the absence of the hepcidin antagonist fursultiamine than in the presence of fursultiamine (Figure 4B). The same results were found in the BMVECs stimulated by C57BL/6 astrocyte supernatant compared with Hepc−/− astrocyte supernatant (Figure 4C). These results suggest that increased hepcidin may prevent accumulative brain iron flow into circulation by inhibiting BMVECs intracellular iron efflux.
Increased Serum Hepcidin Involved in Brain Iron Accumulation After ICH
We previously showed that serum hepcidin contents are positively correlated with the outcome of patients with ICH17 and that serum hepcidin was significantly increased in mouse ICH models. Therefore, we investigated the roles of serum hepcidin in brain iron metabolism by using the parabiosis model and constructing C57BL/6-C57BL/6, C57BL/6-Hepc−/−, and Hepc−/−-Hepc−/− parabiosis ICH models after 20 days of parabiosis (Figure 5A). The parabiosis models share a similar blood after 10 to 14 days of parabiosis,24 and the periphery hepcidin in C57BL/6 mice can enter into the circulation of Hepc−/− mice (Figure 5B). To further explore whether the serum hepcidin could enter into brain tissues to become involved in brain iron metabolism after ICH, we found that the hepcidin did not exist in brain tissue of heterochronic Hepc−/− ICH mice detected by Western blot (online-only Data Supplement Figure VA) and immunofluorescent staining (online-only Data Supplement Figure VB). Brain iron levels peaked 14 days after ICH, at which time point we measured the brain iron levels. The brain iron contents were significantly higher in heterochronic Hepc−/− mice exposed to C57BL/6 circulation compared with Hepc−/− mice exposed to Hepc−/− circulation (Figure 5C). These results show that increased periphery hepcidin is also involved in promoting brain iron accumulation after ICH.
TLR4/MyD88 Signaling Pathway Upregulated Hepcidin Expression After ICH
IL-6/STAT3 signaling has been shown to promote hepcidin expression,19 and IL-6 is one of the proinflammatory cytokines of TLR4 signaling after ICH.21 Therefore, we investigated the roles of TLR4 signaling in hepcidin expression after ICH and found that TLR4−/− and MyD88−/− mice showed pronounced decreases in hepcidin content in the brain at day 3 after ICH compared with C57BL/6 mice, whereas TRIF−/− mice showed no significant difference (Figure 6A). Furthermore, the results of immunofluorescent staining also showed that double- (hepcidin-25 and GFAP) positive cells were lower in TLR4−/− mice than in C57BL/6 mice at day 3 after ICH (Figure 6B). In addition, we found that IL-6 levels in the brain were increased after ICH, peaking at day 3 (Figure 6C). Moreover, we found that brain IL-6 and phosphorylated STAT3 levels were decreased in TLR4−/− and MyD88−/− mice in contrast to C57BL/6 mice, whereas TRIF−/− mice were similar to C57BL/6 mice (Figure 6D). These results show that the TLR4/MyD88 signaling pathway increased IL-6 production, which further activated STAT3 phosphorylation to induce hepcidin expression after ICH.
To further confirm the effects of TLR4/MyD88 signaling on brain iron accumulation, we compared serum and brain iron contents among TLR4−/−, MyD88−/−, TRIF−/−, and C57BL/6 mouse ICH models. Our findings showed that TLR4−/− (Figure 6E) and MyD88−/− (online-only Data Supplement Figure VIA) mice exhibited lower serum iron levels than C57BL/6 mice, whereas there was no significant difference in C57BL/6 and TRIF−/− mice (online-only Data Supplement Figure VIB). In addition, we found no significant difference in brain iron levels among TLR4−/−, MyD88−/−, TRIF−/−, and C57BL/6 normal mice (online-only Data Supplement Figure VIC). Then, brain iron contents were measured after ICH and were found to be lower in TLR4−/− mice compared with C57BL/6 mice at 7, 14, and 28 days after ICH, whereas there was no significant difference in the acute phase of ICH (Figure 6F). Additionally, brain iron levels in TLR4−/− mice peaked at day 7 after ICH (Figure 6F). MyD88−/− mice also exhibited a tendency in brain iron content similar to that of the TLR4−/− mice (online-only Data Supplement Figure VID), whereas the results of the TRIF−/− mice were consistent with those of the C57BL/6 mice after ICH (online-only Data Supplement Figure VIE).
Next, we further investigated TLR4 signaling in hepcidin expression and its effects in vitro. First, we found that the supernatant hepcidin contents of hemoglobin-stimulated TLR4−/− astrocytes were lower than the C57BL/6 astrocytes (Figure 7A). In addition, the remaining intracellular ferritin levels in BMVECs stimulated by C57BL/6 astrocyte supernatants were higher than that stimulated by TLR4−/− astrocyte supernatants (Figure 7B). These results suggest that TLR4/MyD88 signaling increased hepcidin expression, which prevents brain iron evacuation via inhibition of the intracellular iron of BMVECs entering into circulation after ICH.
TAK242 Promoted Brain Iron Efflux and Improved Cognition Impairment After ICH
We have previously shown that the TLR4 antagonist TAK242 significantly improved neurological deficits in the acute phase of ICH.28 Here, we used TAK242 to evaluate the therapeutic effects of brain iron on targeting TLR4-mediated inflammation. Consecutive injection of TAK242 for 5 days significantly reduced brain iron accumulation at days 14 and 28 after ICH, whereas there was no significant difference 7 days after ICH (Figure 8A). We subsequently confirmed iron-related impairment in the later phase of ICH during the application of TAK242 using the RAWM. Our findings on TAK242 enhancement of learning and memory for hidden platform location compared with vehicle groups during the testing phase (Figure 8B) suggest that targeting TLR4-mediated inflammation not only may reduce acute phase brain injury but also may improve later-phase cognition impairment that may be related partly to the reduction of brain iron accumulation after ICH.
In the present study, we provide evidence of the effects of increased hepcidin caused by TLR4 signaling on preventing brain iron entering into circulation across the blood-brain barrier and aggravating oxidative brain injury and cognition impairment in ICH mice (online-only Data Supplement Figure VII). Using the parabiosis ICH models, we revealed for the first time that increased periphery hepcidin also has a significant effect on promoting brain iron accumulation after ICH, which may be attributed to the increased hepcidin caused by TLR4 signaling pathway preventing BMVECs intracellular iron efflux. Accordingly, inhibiting inflammation exhibited therapeutic effects of reduced oxidative brain injury and improvement of cognition impairment after ICH.
The research results of the cellular origin of hepcidin in the neuroinflammatory brain are inconsistent. In our study, astrocytes were the major source of hepcidin production in brain tissues after ICH, and neurons did not produce hepcidin, which is consistent with Urrutia et al’s36 report that astrocytes are also the major source of hepcidin under the condition of inflammatory cytokine stimulation. However, in another study, a significant increase in hepcidin expression was found in neurons within the cortex and substantia nigra caused by lipopolysaccharide.37 Although neurons do not respond to lipopolysaccharide, the increase in hepcidin production in neurons is dependent on the microglia.37 In addition, despite the increased hepcidin observed in the ischemic brain, no information about the cellular source of hepcidin was reported.38 The inconsistent results on the source of hepcidin may be due to different experimental models, conditions, or other unknown factors that may influence hepcidin expression in brain tissue.
In the present study, we found that increased hepcidin encircled the BMVECs (Figure 1E, red arrow), suggesting that hepcidin plays an important role in regulating brain iron metabolism by influencing the transference of iron across the blood-brain barrier. In the Hepc−/− mice, we found that brain iron evacuation was increased, whereas giving a supplement of human hepcidin-25 peptide aggravated brain iron accumulation. In addition, data obtained from the in vitro study showed that increased hepcidin inhibits BMVEC intracellular iron efflux, maybe because hepcidin binds to ferroportin and induces its internalization, resulting in a decrease in the cellular iron efflux channel.16,39 These results suggest that increased hepcidin prevents brain iron from evacuating into circulation after ICH. Because the hepcidin antagonist fursultiamine has been shown only to have antagonistic effects on hepcidin in the in vitro study,40 we did not investigate the function of fursultiamine in our in vivo experimental ICH models. We previously showed that serum hepcidin contributes to poor outcomes in patients with ICH17 and found that serum hepcidin was increased in mice after ICH. Therefore, to demonstrate that serum hepcidin is also involved in brain iron accumulation after ICH, we used the parabiosis ICH models and found that brain iron content at day 14 after ICH in heterochronic Hepc−/− mice was higher than in isochronic Hepc−/− mice because heterochronic Hepc−/− mice received hepcidin from the heterochronic C57BL/6 mice via an established blood communication. In addition, the result that serum hepcidin cannot be detected in the brain tissues of Hepc−/− mice was consistent with other results that hepcidin would be degraded by the lysosome when taken into the cell.16
Accumulative brain iron can lead to oxidative brain injury2 in the acute phase of ICH and cognition impairment7,41,42 in the later phase of ICH. We measured the reactive oxygen species index of brain DCFH and 8-isoprostane levels to evaluate oxidative brain injury in Hepc−/− ICH mice because lower brain iron content was found in Hepc−/− mice than in C57BL/6 mice. An improvement in brain water content and neurological deficient score at 3, 5, and 7 days after ICH was also found in Hepc−/− mice. To further investigate the long-term effect of reduced brain iron after ICH, we used the RAWM to evaluate the cognition impairment of mice at day 28 after ICH and found that Hepc−/− mice exhibited more improvement in spatial learning and memory than C57BL/6 mice. Furthermore, these effects in C57BL/6 mice were aggravated by the intravenous injection of human hepcidin-25 peptide after ICH. These results suggest that increased hepcidin in both brain and serum results in the inhibition of brain iron evacuation that causes iron-related brain injury after ICH.
However, in the brain iron–overload rats, hepcidin has been shown to suppress brain iron accumulation by inhibiting the transport of transferrin-bound iron from the periphery into the brain,43 and pretreatment of rats with adenovirus hepcidin prevents iron-induced oxidative stress in the brain.44 The other study showed that increased hepcidin caused by inflammation was connected to brain iron accumulation,36 which was consistent with our results. The contradictory effects of hepcidin on brain iron accumulation may be due to different experimental models or conditions. In our opinion, accumulated iron in the brain already exists in the iron-overloaded brain, and only the promotion of brain iron entering into circulation can solve this issue.
Increased hepcidin is connected to brain iron accumulation after ICH; thus, elucidating the mechanism underlying hepcidin expression seems to be important for ICH treatment. The IL-6/STAT3 signaling pathway has been demonstrated to be an important contributor to hepcidin expression.19 Therefore, in this study, we further investigated the brain levels of IL-6 and found that the levels of IL-6 also peaked at day 3 after ICH. We previously determined IL-6 to be the downstream inflammatory cytokine of the TLR4 signaling pathway after ICH.21 To further explore the role of TLR4 signaling in hepcidin expression after ICH, using the TLR4−/− mice, we found that the brain levels of hepcidin were decreased compared with C57BL/6 mice at day 3. Similarly, the IL-6 and phosphorylated STAT3 levels were also decreased in TLR4−/− ICH mice at day 3 after ICH. Moreover, we investigated which pathway of TLR4 signaling was involved in promoting hepcidin expression. The results of the Western blot showed that brain hepcidin, IL-6, and phosphorylated STAT3 expression at 3 days after ICH was also decreased in MyD88−/− mice, with no significant difference in TRIF−/− mice compared with C57BL/6 mice. These results suggest that the TLR4/MyD88 signaling pathway promotes hepcidin expression via the IL-6/STAT3 pathway. In addition, the TLR4−/− and MyD88−/− mice exhibited lower brain iron accumulation at days 7, 14, and 28 after ICH, whereas the levels of brain iron content in TLR4−/− and MyD88−/− mice peaked sooner than in C57BL/6 mice. The reason may be that TLR4−/− and MyD88−/− mice have faster hematoma absorption than C57BL/6 mice,27 or there may be other factors that are influenced by inflammation to counteract the effects of reduced hepcidin on brain iron evacuation in the acute phase of ICH.
Our results showed that increased hepcidin caused byTLR4/MyD88 signaling prevents brain iron efflux into circulation after ICH. The presented insight into increased hepcidin expression caused by inflammation after ICH identifies a mechanistic target for muting inflammation to promote brain iron efflux in iron-related brain injuries after ICH.
The authors thank Prof. Frederick Colbourne (Department of Psychology and Center for Neuroscience, University of Alberta, Edmonton, AB, Canada) for providing the brain nonheme iron measurement protocol.
Sources of Funding
This work was supported by grants from the National Natural Science Foundation of China (81471191 and 81601028), the National Natural Science Fund for Distinguished Young Scholars (81525008), and the National Basic Research Program of China (973 Program) (2014CB541605).
Sources of Funding, see page 1037
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.116.021881/-/DC1.
Circulation is available at http://circ.ahajournals.org.
- Received February 4, 2016.
- Accepted July 27, 2016.
- © 2016 American Heart Association, Inc.
- Wu H,
- Wu T,
- Xu X,
- Wang J,
- Wang J
- Qin Y,
- Zhu W,
- Zhan C,
- Zhao L,
- Wang J,
- Tian Q,
- Wang W
- Huang XT,
- Qian ZM,
- He X,
- Gong Q,
- Wu KC,
- Jiang LR,
- Lu LN,
- Zhu ZJ,
- Zhang HY,
- Yung WH,
- Ke Y
- Xie Q,
- Gu Y,
- Hua Y,
- Liu W,
- Keep RF,
- Xi G
- Selim M,
- Yeatts S,
- Goldstein JN,
- Gomes J,
- Greenberg S,
- Morgenstern LB,
- Schlaug G,
- Torbey M,
- Waldman B,
- Xi G,
- Palesch Y
- 14.↵ClinicalTrials.gov. Intracerebral Hemorrhage Deferoxamine Trial-iDEF Trial. https://www.clinicaltrials.gov/ct2/show/NCT02175225. Accessed September 16, 2016.
- Nemeth E,
- Tuttle MS,
- Powelson J,
- Vaughn MB,
- Donovan A,
- Ward DM,
- Ganz T,
- Kaplan J
- Wrighting DM,
- Andrews NC
- Zhao F,
- Hua Y,
- He Y,
- Keep RF,
- Xi G
- 25.↵National Centre for the Replacement Refinement & Reduction of Animals in Research. ARRIVE Guidelines. http://www.nc3rs.org.uk/arrive-guidelines. Accessed September 16, 2016.
- Fang H,
- Chen J,
- Lin S,
- Wang P,
- Wang Y,
- Xiong X,
- Yang Q
- Wang YC,
- Wang PF,
- Fang H,
- Chen J,
- Xiong XY,
- Yang QW
- Boscolo A,
- Starr JA,
- Sanchez V,
- Lunardi N,
- DiGruccio MR,
- Ori C,
- Erisir A,
- Trimmer P,
- Bennett J,
- Jevtovic-Todorovic V
- Fung E,
- Sugianto P,
- Hsu J,
- Damoiseaux R,
- Ganz T,
- Nemeth E
- Garcia PY,
- Roussel M,
- Bugnicourt JM,
- Lamy C,
- Canaple S,
- Peltier J,
- Loas G,
- Deramond H,
- Godefroy O
- Benedictus MR,
- Hochart A,
- Rossi C,
- Boulouis G,
- Hénon H,
- van der Flier WM,
- Cordonnier C
What Is New?
Using parabiosis and intracerebral hemorrhage surgery, this study showed that the Toll-like receptor 4/MyD88 signaling pathway increased both brain astrocyte-derived and serum hepcidin, which resulted in the inhibition of brain iron efflux into circulation after intracerebral hemorrhage via binding to the brain ferroportin of microvascular endothelial cells, which reduces iron efflux channels.
What Are the Clinical Implications?
Intracerebral hemorrhage induced inflammatory responses that promote hepcidin expression, suggesting that muting inflammation or interfering with increased hepcidin may be a therapeutic target for intracerebral hemorrhage treatment to promote brain iron efflux and to reduce the cause of brain injuries and cognition impairments.