Dysregulation of the (immuno)proteasome pathway in malformations of cortical development

The cases included in this study were obtained from the archives of the Departments of Neuropathology of the Academic Medical Center (AMC, University of Amsterdam, The Netherlands), the University Medical Center Utrecht (UMCU, The Netherlands), and the Medical University Vienna (MUV, Austria). A total of 23 brain tissue specimens, removed from patients undergoing surgery for intractable epilepsy, were examined. The tissue was obtained and used in accordance with the Declaration of Helsinki and the AMC Research Code provided by the Medical Ethics Committee and approved by the committee of the UMCU Biobank. This study was also approved by the Ethical Committee of the Medical University of Vienna. All cases were reviewed independently by two neuropathologists, and the diagnosis of FCD was confirmed according to the international consensus classification system recently proposed for grading FCD [21]. All patients with cortical tubers fulfilled the diagnostic criteria for TSC [22]. None of the FCD patients fulfilled the diagnostic criteria for TSC. Table 1 summarizes the clinical findings of patients with MCD and epilepsy (6 mild MCD (mMCD), 5 FCD IIa, 6 FCD IIb, 6 TSC tubers: 4 TSC2/2 TSC1; pre-operative seizure frequency/month, mean ± SEM: mMCD 19.8 ± 6.7; FCD II 149 ± 68.7; TSC 114.8 ± 24.2); seizure frequencies were recorded (video-electroencephalographic monitoring) at the time of the preoperative evaluation. One tuber specimen was obtained postmortem (age 32 years; male; TSC2). Hippocampal specimens from patients with Alzheimer’s disease (AD; n = 4; 3 females and 1 male; Braak stages V and VI, age 81.7 ± 2.8) were also examined as positive controls. In addition, normal-appearing control cortex and white matter were obtained at autopsy from six young adult control patients (Table 1), without history of seizures or other neurological diseases. All autopsies were performed within 24 h after death.

Brain tissue from control and MCD patients was fixed in 10 % buffered formalin and embedded in paraffin. Paraffin-embedded tissue was sectioned at 5 μm, mounted on pre-coated glass slides (Star Frost, Waldemar Knittel GmbH, Braunschweig, Germany), and used for histology and immunohistochemistry. One representative paraffin block per case was sectioned, stained, and assessed. Sections were processed for hematoxylin eosin stainings, as well as for immunohistochemical stainings for a number of neuronal and glial markers and antibodies against the constitutive (β1, β5) and immunoproteasome (β1i, β5i) subunits (Table 2). These antibodies have been extensively tested on human liver and brain tissues [23, 24], including surgical brain specimens from patients with mesial temporal lobe epilepsy revealing bands at the expected molecular weight ([17]; Additional file 1: Figure S1). To detect differences in labeling related to technical variables such as tissue fixation, we also tested the antibodies in specimens of selected regions (temporal cortex/hippocampus) collected at autopsy and immediately fixed in formalin for 24 h (same fixation time used for the surgical specimens); no differences in the immunoreactivity pattern were observed.

Single-label immunohistochemistry was performed as previously described [25]. Sections were deparaffinated in xylene, rinsed in ethanol (100, 95, and 70 %) and incubated for 20 min in 0.3 % hydrogen peroxide diluted in methanol. Antigen retrieval was performed using a pressure cooker in 0.1 M citrate buffer pH 6.0 at 120 °C for 10 min. Slides were washed with phosphate-buffered saline (PBS; 0.1 M, pH 7.4) and incubated overnight with the primary antibody in PBS at 4 °C. After washing in PBS, sections were stained with a polymer-based peroxidase immunohistochemistry detection kit (PowerVision Peroxidase System, ImmunoVision, Brisbane, CA, USA). The 3,3′-diaminobenzidine tetrahydrochloride was used as chromogen. Sections were dehydrated in alcohol and xylene and coverslipped.

Double-labeling of β1, β1i, β5, or β5i with NeuN (neuronal nuclear protein (NeuN; mouse clone MAB377; Chemicon, Temecula, CA, USA; 1:2000), GFAP (polyclonal rabbit, DAKO, Glostrup, Denmark; 1:4000, or monoclonal mouse, Sigma-Aldrich, St. Louis, MO, USA; 1:4000), HLA-I (mouse clone HC-10, 1:200), or HLA-II (mouse anti-human leukocyte antigen (HLA)-DP, DQ, DR, mouse clone CR3/43; DAKO; 1:400) was performed as previously described [26]). Sections were incubated with BrightVision poly-alkaline phosphatase (AP)-anti-rabbit or anti-mouse (Immunologic, Duiven, The Netherlands) for 30 min at room temperature and washed with PBS. AP activity was visualized with the AP substrate kit III Vector Blue (SK-5300, Vector Laboratories Inc., CA, USA). To remove the first primary antibody, sections were incubated at 121 °C in citrate buffer (10 mM NaCi, pH 6.0) for 10 min. Incubation with the second primary antibody was performed overnight at 4 °C. Sections with primary antibody other than rabbit were incubated with post-antibody blocking from the BrightVision+ system (containing rabbit-α-mouse IgG; Immunologic, Duiven, The Netherlands). AP activity was visualized with the alkaline phosphatase substrate kit I Vector Red (SK-5100; Vector Laboratories Inc., CA, USA). Sections incubated without the primary antibody, with preimmune sera, or with the antibody preincubated with the antigenic peptide (for the polyclonal β5) were essentially blank.

All labeled tissue sections were evaluated by two independent observers for the presence or absence of various histopathological parameters and specific immunoreactivity (IR) for the different markers used for the diagnosis of mMCD, FCD subtypes, and TSC tubers. We also semi-quantitatively evaluated the IR (nucleus and cytoplasm in glial and neuronal cells) of β1, β1i, β5, and β5i. The intensity of the staining was evaluated using a scale of 0–3 (0: no; 1: weak; 2: moderate; 3: strong staining). All areas of the lesion were examined, and the score represents the predominant cell staining intensity found in each case. The frequency of β1, β1i, β5, or β5i positive cells ((1) rare; (2) sparse; (3) high) was also evaluated to give information about the relative number of positive cells within the lesion. We also evaluated intensity and frequency of pS6 and IL-1β staining. As described in previous studies [25, 27], the product of the intensity and frequency scores was taken to give the overall score (total score; immunoreactivity score (IRS), Table 3). Quantification of signal intensity using ImageJ software was performed for β1i and β5i subunits (Additional file 2: Figure S2).

Primary fetal astrocyte-enriched cell cultures were obtained from human fetal brain tissue (14–19 weeks of gestation) obtained from the HIS-Mouse (human immune system mouse) facility of the AMC, Amsterdam. All materials have been collected from donors from whom a written informed consent for the use of the material for research purposes had been obtained by the Bloemenhove Clinic (Heemstede, The Netherlands); these informed consents are kept together with the medical record of the donor by the clinic. The tissue was obtained in accordance with the Declaration of Helsinki and the AMC Research Code provided by the Medical Ethics Committee of the AMC. Cell isolation was performed as described elsewhere [28‐30]. Briefly, after the removal of the blood vessels, the tissue was mechanically minced into smaller fragments and enzymatically digested by incubating at 37 °C for 30 min with 2.5 % trypsin (Sigma-Aldrich; St. Louis, MO, USA). The tissue was washed with incubation medium containing Dulbecco’s modified Eagle’s medium (DMEM)/HAM F10 (1:1) medium (Gibco, Life Technologies, Grand Island, New York, USA), supplemented with 50 units/ml penicillin, 50 μg/ml streptomycin, and 10 % fetal calf serum (FCS; Gibco, Life Technologies, Grand Island, New York, USA) and triturated by passing through a 70 μm mesh filter. Cell suspension was incubated at 37 °C, 5 % CO₂ for 48 h to let glial cells adhere to the culture flask before it was washed with PBS to remove excess of myelin and cell debris. Cultures were subsequently refreshed twice a week. Cultures reached confluence after 2–3 weeks.

Primary FCD astrocyte cultures were derived from a surgical human brain specimen obtained from a patient with FCD type IIA (age at surgery, 16 years; female; location, frontal; seizure frequency, thrice per week; duration of epilepsy, 11 years) undergoing epilepsy surgery at the Department of Pediatrics/Neurosurgery of the Medical University Vienna (Vienna, Austria). FCD astrocyte cultures were established in the same manner as described above for fetal cultures.

Secondary astrocyte cultures for experimental manipulation were established by trypsinizing confluent cultures and sub-plating onto poly-l-lysine (PLL; 15 μg/ml, Sigma-Aldrich)-precoated 12- and 24-well plates (Costar, Cambridge, MA, USA; 5 × 10⁴ cells/well in a 12-well plate for RNA isolation and PCR; 2.5 × 10⁴ cells/well for immunocytochemistry). In the present study, astrocytes were used for analyses at passages 2–4.

Cell cultures were stimulated with human recombinant (r)IL-1 β (PeproTech, Rocky Hill, NJ, USA; 10 ng/ml) or in some experiments with lipopolysaccharide (LPS; 100 ng/ml; Sigma-Aldrich, St. Louis, USA) for 24 h. Treatment of FCD-derived astrocytes with rapamycin (100 nM) was started 24 h before and continued during IL-1β stimulation. Cells were harvested 24 h after stimulation. Viability of human cell cultures was not influenced by the performed treatments (Additional file 3: Figure S3).

For immunofluorescent staining of cell cultures, sections were incubated with the primary antibodies for β1, β1i, β5, or β5i for 1 h at RT, followed by 2 h of incubation at RT with Alexa Fluor® 568-conjugated anti-rabbit or Alexa Fluor® 488-conjugated anti-mouse IgG (1:200, Molecular Probes, The Netherlands) together with Alexa Fluor® 488 or 594 Phalloidin (1:200, Molecular Probes, Plaats, The Netherlands) for counterstaining actin filaments. Sections were mounted using VECTASHIELD with DAPI (Vector Laboratories Inc., Burlingame, CA, USA). Fluorescent microscopy was performed using a Leica Confocal Microscope TSC SP-8X (Leica, Son, the Netherlands) at ×40 magnification (bidirectional X, speed 600 Hz, pinhole 1.00 AU).

For RNA isolation, cell culture material was homogenized in Qiazol Lysis Reagent (Qiagen Benelux, Venlo, The Netherlands). Total RNA was isolated using the miRNeasy Mini kit (Qiagen Benelux, Venlo, The Netherlands) according to the manufacturer’s instructions. The concentration and purity of RNA were determined at 260/280 nm using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). To evaluate β1, β1i, β5 or β5i, and IFNγ mRNA expression, 200 ng of cell-culture-derived total RNA was reverse-transcribed into cDNA using oligo dT primers. PCRs were run on a Roche LightCycler 480 thermocycler (Roche Applied Science, Basel, Switzerland) using the following primers: β1 (forward: accagctcggtttccaca, reverse: cccggtatcggtaacacatc); β5 (forward: gagtctcagtgatggtctgagc, reverse: actccatggcggaacttg); β1i (forward: accaaccggggacttacc, reverse: tcaaacactcggttcaccac); β5i (forward: ccctacccacccctgttt; reverse: cacccagggactggaaga); and IFN-γ (forward: gcaagatcccatgggttgtgt; reverse: ctggctcagattgcaggcata). Quantification of data was performed using the computer program LinRegPCR in which linear regression on the log (fluorescence) per cycle number data is applied to determine the amplification efficiency per sample [31, 32]. The starting concentration of each specific product was divided by the geometric mean of the starting concentration of the reference genes (EF1α and C1orf43), and this ratio was compared between groups.

Statistical analyses were performed with GraphPad Prism software (Graphpad Software Inc., La Jolla, CA, USA). To assess differences in immunoreactivity score between multiple groups, non-parametric Kruskal-Wallis followed with Mann-Whitney U test was used. Correlations were assessed using Spearman’s (rho) rank correlation test. For cell culture data, Mann-Whitney U test was used to asses differences between different conditions. P<0.05 was assumed to indicate a significant difference.

The clinical features of the cases included in this study are summarized in Table 1. All operated patients had a history of chronic pharmacoresistant epilepsy. In this study, we included patients with mild degree of cortical dysplasia (mMCDs; [33]). Age at surgery, seizure duration, and seizure frequency were not statistically different between patients with FCD II and mMCD in this cohort, as well as between the FCD IIa and FCD IIb cases included in our cohort. Accordingly to the international consensus classification system of FCD [21], FCD II represents isolated focal lesions with architectural and dysmorphic abnormalities (FCD IIa with dysmorphic neurons only; FCD IIb with dysmorphic neurons and balloon cells; Figs. 1, 2, 3, and 4c–e). TSC patients were younger compared to mMCD and FCD patients. All six TSC tubers displayed similar histopathological features, including loss of lamination, astrogliosis, dysmorphic neurons, and giant cells with pale eosinophilic cytoplasm ([34]; Figs. 1, 2, 3, and 4f, g).

Expression of β1, β1i, β5, and β5i was observed in FCD, cortical tubers, and mMCD specimens (Figs. 1, 2, 3, and 4; Additional file 2: Figure S2 and Additional file 4: Figure S4). We observed differences in the expression level as well as in the cell-specific and subcellular distribution of the different subunits (Table 3).

Our data show prominent expression of both constitutive and immunoproteasome subunits in MCD, such as FCD and TSC, associated with the mTOR pathway. In all the FCD II and TSC specimens examined, the IR for β1, β1, β5, and β5i was increased within the dysplastic regions where prominent gliosis and the presence of dysmorphic neurons and balloon or giant cells (in FCD IIb and TSC, respectively) was observed. Constitutive and particularly immunoproteasome subunits displayed increased expression compared to control but also compared to mMCD specimens from patients with chronic epilepsy. These results indicate that increased expression of proteasome subunits is not simply an effect of seizure activity; moreover, the duration of epilepsy in mMCD cases did not differ from FCDs and was even longer compared to TSC cases. However, a positive correlation was observed between nuclear glial and neuronal proteasome subunit expression and the pre-operative seizure frequency. We acknowledge limitations to the interpretation of these results; therefore, an evaluation of the real biological contribution of proteasome subunit expression to seizure generation and frequency deserves further investigation in experimental models.

Several proteasome subunits show nuclear localization signaling [37], and previous studies in the human brain indicate that proteasomes are expressed in both cytoplasm and nuclei of different cell types, including glial and neuronal cells [24, 38]. Immunoproteasome expression restricted to nuclei of astrocytes has been reported in the brain after an infection with lymphocytic choriomeningitis virus, suggesting involvement of the nuclear envelope in the compartmentalization of immature proteasome precursors [39]. Whether the nuclear proteasome subunits represent (as suggested by Kremer et al. [39]) immature proteasome precursors or are proteolytically active remains still to be investigated. The nuclear proteasome subunit accumulation may reflect the induction of the proteasome system under conditions associated with cell injury and inflammation with the possibility of nucleo-cytoplasmic transfer in cells, as glial cells, undergoing cell division or during apoptosis [37]. However, the β1i subunit in the nuclear-enriched fraction has also been detected in its catalytically active form [40], and several studies indicate a possible functional role of the immunoproteasome in transcriptional regulation [41‐43]. The expression pattern, either nuclear or cytoplasmic proteasome expression, can be influenced by the type and duration of fixation [37]. However, similar pattern was observed in surgical and postmortem TSC brain tissue.

One of the major regulatory factors of immunoproteasome induction is inflammation [43, 44]. Several studies confirmed the occurrence of complex inflammatory changes, involving both glial and neuronal cells, and the activation of the IL-1β pathway, particularly in FCD II and TSC [20, 34, 35, 45‐48]. Thus, the pro-inflammatory environment may contribute to the activation of the proteasome system, particularly to the induction and expression of the immunoproteasome subunits. Accordingly, our in vitro studies in human astrocytes and FCD cultures indicate that IL-1β treatment increases the induction of, in particular, the immunoproteasome subunits β1i and β5i, with the increase of their perinuclear-nuclear localization. This observation supports the role of astrocytes as targets of regulation of the immunoproteasome under various conditions associated with the activation of the IL-1β pathway [16] and indicates that pro-inflammatory cytokines, other than IFNγ, may regulate immunoproteasome expression. Activation of inflammatory pathways, including IL-1β, may also play a role in the regulation of immunoproteasome expression in other cell types, such as neurons. Accordingly, we found a positive correlation between the expression of immunoproteasome subunits in both glial and neuronal cells and the expression of IL-1β within the dysplastic area in FCD II and in TSC specimens. Moreover, increasing evidence supports the role of the immunoproteasome in the activation of the NF-kB pathway, modulation of pro-inflammatory cytokine production, and oxidative stress response [9, 43, 49‐52]. Induction of the β5i subunit has also been shown in vivo following activation of the Toll-like receptor 4 (TLR4)-mediated NF-kB signaling pathway by LPS [53]. Thus, we may speculate about the existence of a reinforcing feedback loop between NF-kB pathway and the immunoproteasome system, which may play a crucial role in perpetuating the pro-epileptogenic inflammatory response in epilepsy. Interestingly, Mishto et al. [18] provide additional experimental evidence of the regulation of β5i subunit by TLR4 signaling in epileptogenic tissue.

The immunoproteasome is known to improve MHC class I (MHC-I) antigen presentation and has been suggested to have a central function at the interface between the innate and adaptive immune system (reviewed in [11]). Interestingly, FCD II and TSC specimens are characterized by prominent activation of both innate and adaptive immune responses (for review, see [20, 36]). Moreover, recent studies provide evidence of an upregulation of MHC-I, involving also balloons/giant cells and neurons, in both FCD II and TSC specimens [54].

FCD II and TSC cases are characterized by architectural or cellular changes associated with mTOR pathway activation [20, 21]. The innate and adaptive immune responses have also been shown to be influenced by the mTOR pathway [55‐57]. Moreover, the mTOR complex 1 (mTORC1) has been identified as a key regulator of autophagy [58, 59], a pathway which is defective in FCD II and TSC [60]. Increasing evidence indicates a strong relationship with tight coordination between the autophagy and the proteasome system [61]. Thus, we cannot exclude a role of mTOR in the regulation of the proteasome system, including immunoproteasome subunit expression. Accordingly, we observed a positive correlation between immunoproteasome subunit expression in neurons and pS6 expression, indicating the activation of the mTOR signal transduction pathway. The relationship between mTOR and proteasome system is also supported by the in vitro experiments showing that inhibition of the mTOR pathway by the potent allosteric mTORC1 inhibitor rapamycin was able to reduce the level of expression of inducible proteasome subunits in FCD-derived cells. This is in agreement with a recent study showing reduction of the immunoproteasome by rapamycin in H9c2 cells as well as in mouse heart in vivo [62]. Evaluation of the possible effect of rapamycin on the expression of the brain immunoproteasome in vivo deserves further studies and is presently under investigation [63].

An example of the possible use of inhibition of the immunoproteasome as therapeutic strategy in epilepsy is represented by the study of Mishto and colleagues [18] in which specific inhibition of the β5i subunit by ONX-0914 [64] resulted in prevention, or significant delay, of 4-aminopyridine-induced seizure-like events in acute rat hippocampal/entorhinal cortex slices, particularly in slices of epileptic rats. Clinically approved proteasome inhibitors targeting the catalytic activity of both the constitutive proteasome and the immunoproteasome have been already used in hematological malignancies [65‐67]. New-generation small molecules specifically targeting the immunoproteasome are under clinical development and have been already evaluated in a large variety of animal models of autoimmune diseases and proposed as novel therapeutic approaches for patient with multiple sclerosis, as well as in neurodegenerative diseases (for reviews, see [16, 68, 69]).

However, recently alternative functions for the immunoproteasome have also been considered, suggesting that the induction of the immunoproteasome may also play a role in neuronal protection and repair after injury, contributing to the preservation of cell viability upon cytokine-induced oxidative stress [49, 70, 71], which is known to be increased within the TSC tubers [72]. In particular, evidence has been provided that the immunoproteasome plays a role in the clearance of damaged proteins accumulating upon inflammation or oxidative stress (for review, see [49]), which are also detected in TSC and FCD [73]. Accordingly, the formation of aggresome-like-induced structures and increased sensitivity to apoptosis has been reported in immunoproteasome deficiency in cells and in a murine inflammation model [49, 71]. Additional studies support alternative physiological function of the immunoproteasome subunits, including also cell proliferation, cell signaling, and synaptic remodeling (for review, see [49, 74, 75]). Thus, an effective therapeutic intervention based on the immunoproteasome has to take into consideration the preservation of the potential beneficial functions of its activation, particularly during brain development.