Keywords

Introduction

Haptoglobin (Hp) is an endogenous hemoglobin (Hb)-binding protein present in blood plasma. In blood, Hp binds cell-free Hb (derived from damaged or senescent erythrocytes) to form highly stable Hb-Hp complexes [1]. This process is capable of shielding and stabilizing the heme iron within the Hb hydrophobic pocket and thereby blocking the pro-oxidative property of Hb [2, 3]. Normally, Hp is produced by hepatocytes and then released to the blood [4, 5] where it binds the cell-free Hb and mediates the fast removal of Hb from circulation [6, 7]. The clearance of the Hp-Hb complexes occurs via circulating blood monocytes and tissue macrophages that express the Hb scavenger receptor CD163 [810]. In addition, the uptake of Hp-Hb complexes by hepatocytes plays an important role in iron recycling. Hp indirectly exerts a broad range of anti-inflammatory activities and plays an anti-oxidant role most notably by virtue of its ability to bind free Hb, accelerating the rapid clearance of Hb by monocytes and macrophages.

The Hp reservoir in blood plasma is relatively high, allowing for efficient buffering of the Hb in case of acute hemolytic events. Although the role of Hp in intravascular hemolysis is well accepted, the role of Hp in the brain is essentially unknown [11]. Limited studies suggest the presence of Hp in the neural retina [12] and subarachnoid blood clot [13]; however, its expression profile and role in the brain is not clear.

Following ICH, a large amount of Hb is released into the brain parenchyma after erythrocytes undergo hemolysis within the hematoma [14, 15]. The cell-free Hb is a potent neurotoxin capable of initiating deleterious reactions by causing oxidative damage to lipids, DNA and proteins [16, 17]; caspase activation [18, 19]; blood-brain-barrier (BBB) disruption [14, 15, 20]; and irreversible neuronal damage [2123]. We postulate that Hp may benefit the brain after ICH assuming Hp can either penetrate to the brain through the disrupted BBB or being locally produced by the brain cells. Here, we review our finding [24] regarding the expression, localization and biological function of Hp in the brain after ICH.

Results

Hp Is Increased in the Brain After ICH: The Role of Oligodendrocytes

To explore the role of Hp in the brain after ICH, we first tested the expression of Hp protein in the brain using Western blot analysis. In naïve Sprague-Dawley (SD) rats, Hp expression in the brain is very low. However, Hp protein in the brain increases robustly after ICH, which is seen in as soon as 3 h and subsequently increases over time in the first 3 days and remains at higher levels for at least 7 days in the hematoma-affected brain tissues [24]. To verify the origin of brain Hp protein, we measured the Hp mRNA changes in the brain tissue using RT-PCR. We found that Hp mRNA levels share a similar temporal pattern with Hp protein changes, suggesting that Hp is produced locally in the brain. However, we cannot rule out that additional Hp protein may enter the brain through a disrupted BBB.

Next, we studied the spatial distribution of Hp in the brain using immunofluorescence [24]. We found that the Hp signal is strong in the hematoma-affected cortex, corpus callosum and striatum (Fig. 1A, a–c). Using double immunolabeling to identify the specific cell types that express Hp, we found that Hp is primarily co-localized within the MBP-positive oligodendrocytes [24]. Specifically, Hp immunopositive signals localize in the oligodendrocytes’ soma within the grey matter, as well in the fine processes around the myelinated nerve tract in the white matter at the corpus callosum and striatum.

Fig. 1
figure 1_17

(A) The representative Hp immunofluorescence to show Hp localization in the ICH-affected brain. The SD rats were subjected to M-ICH and analyzed for presence of Hp at 24 h using immunofluorescence (now converted to black and white and inverted). Hp immunopositive cells were detected in cerebral cortex (a), corpus callosum (b) and basal ganglia (c) adjacent to the hematoma. Scale bar = 50 μm. ICH: location of ICH. (B) Photograph of Hp localization in the rat primary neuron-glial co-cultures. The oligodendrocytes are visualized with anti-myelin basic protein antibody (MBP, d). The rat Hp is visualized with sheep anti-Hp (e). A merged image of Hp and MBP is shown in (f). The scale bar = 100 μm. Note: Among all of the brain cell types in the field, only MBP-positive cells are immunopositive for Hp. Hp proteins localize in the soma and the fine processing of MBP-positive oligodendrocytes

To further validate the in vivo findings, we studied Hp expression using a primary neuron-glial co-culture system generated from E-18 rat embryos. We found that Hp protein and mRNA are uniquely expressed by MBP-positive oligodendrocytes (Fig. 1B, d–f). Interestingly, the Hp immunopositive signal is detected throughout the soma and fine processes, which is similar to the Hp localizations detected in the rat brains [24].

Using Hp ELISA, we found that Hp is a secretary protein as it can be released by oligodendrocytes into the culture media [24].

Hp Produced by Oligodendrocytes Is Protective to Neurons and Oligodendrocytes

To establish if the Hp produced and secreted by oligodendrocytes is biologically functional, we employed oligodendrocytes cultured from Hp-KO [25] and Hp-Tg [26] mice brains. As expected, at 20 days in culture, Hp-KO oligodendrocytes have practically no Hp presence in the culture media, whereas Hp-Tg oligodendrocytes release high levels of Hp into the culture media (Fig. 2a, b). We then collected the oligodendrocyte-conditioned media and transferred it to the neuron cultures. Compatibility of the media transfer was also confirmed. Next, we measured the injury to the neurons caused by hemolytic products by subjecting the neurons to an “ICH-like” injury (lysed RBC + hypoxia) and assessing the LDH released into the culture media by the injured neurons. At 24 h after adding lysed RBCs, the neurons in the medium from Hp-overexpressing oligodendrocytes demonstrated significantly less injury as compared to the neurons in the medium from Hp-deficient oligodendrocytes (Fig. 2c, d). This suggests that Hp produced by oligodendrocytes can neutralize the hemolytic product-mediated neurotoxicity. In parallel studies, we found that the Hp-Tg oligodendrocytes, similar to neurons, are much more resistant than the Hp-KO oligodendrocytes to the RBC lysate-mediated damage, suggesting that Hp is cytoprotective not only to neurons, but also to oligodendrocytes, the cells that produce Hp [24].

Fig. 2
figure 2_17

Hp produced by oligodendrocytes is cytoprotective. (a) A representative photograph of oligodendrocytes-enriched culture (OEC) prepared from mouse brain at 20 days in vitro. The oligodendrocytes are visualized with MBP immunofluorescence. (b) Bar graph of quantifying Hp protein in the OEC culture media prepared from Hp-KO and Hp-Tg mice, as determined using Hp ELISA. *p ≤ 0.05. (c) A representative photograph of mouse primary cortical neuron cultured for 15 days. The neurons are visualized with MAP2 immunofluorescence. (d) Bar graph of LDH showing the injury to neurons in culture in response to “ICH-like” injury (lysed RBC + hypoxia) in presence of media conditioned by OEC from Hp-KO or Hp-Tg mice. The OEC conditioned media was directly transferred into the neuronal culture by replacing 2/3 volume of the neuronal culture medium and incubated for 15 min before “ICH-like” injury. The LDH in culture media was determined at 24 h. The data are displayed as mean ± SEM (n = 3). *p ≤ 0.05

The Levels of Hp Expression Are Associated with Resistance to ICH-Mediated Brain Damage

To further define the relationship between Hp and ICH, we subjected the Hp-KO, Hp-Tg and WT control mice to a modified ICH injury (M-ICH). In this M-ICH injury model, we injected lysed RBCs (not whole blood) into the basal ganglia to better mimic the events associated with blood toxicity following hemolysis in the hematoma. At 7 days after M-ICH, we assessed brain damage by determining the neurological deficits (NDS; a composite score from a battery of behavioral tests, including foot fault, cylinder, postural reflex and corner tests) [27]. As predicted, Hp-KO mice suffer the most severe neurological deficits, whereas Hp-Tg mice are the least susceptible to M-ICH injury (Fig. 3), demonstrating that Hp in the brain is in fact important in protecting the animals from hemolytic product-mediated brain damage after ICH.

Fig. 3
figure 3_17

Neurological deficit score (NDS) after M-ICH. The Hp-KO, Hp-Tg and WT control mice were subjected to M-ICH, and the neurological deficit was measured at day 7 by a battery of behavioral tests. The data are expressed as mean ± SEM (n = 9) and displayed as percent changes of the NDS in WT mice. *p ≤ 0.05, from all other groups

Conclusion

Hp expression after ICH is increased in the brain, where it is primarily confined to oligodendrocytes, and can be secreted into the extracellular space to play a cytoprotective role against the toxicity of hemolysis products. The temporal and spatial profile of Hp synthesis in the brain appears to be highly strategic for optimal control of hemolysis-mediated cytotoxicity [28, 29]. We propose that Hp could be a potential therapeutic target for ICH.