Introduction

Lafora progressive myoclonus epilepsy (Lafora disease, LD, OMIM 254780, ORPHA501) is a fatal rare autosomal recessive neurodegenerative disorder, which usually occurs during childhood. It is characterized by generalized tonic-clonic seizures, myoclonus, absences, drop attacks, and visual seizures. As the disease progresses, patients present a rapid progressive dementia concomitant with an amplification of seizures, leading to death within a decade after the first symptoms [1, 2]. The hallmark of the disease is the accumulation of insoluble polyglucosan inclusions, called Lafora bodies (LBs), in the cytoplasm of neurons and other cells in peripheral tissues. In the vast majority of patients LD has been associated with mutations in two genes: EPM2A, encoding the glucan phosphatase laforin, and EPM2B, encoding the E3-ubiquitin ligase malin. Although laforin and malin may have independent functions, it has been shown that they form a functional complex in which laforin acts as scaffold and recruits specific substrates to be ubiquitinated by malin. It has also been demonstrated that laforin and malin regulate glycogen synthesis, either by affecting the activity of proteins involved in this process [3–6] or by laforin acting as a glycogen phosphatase and thus being able to dephosphorylate complex glucans [7–9]. This would explain why the absence of laforin or malin leads to aberrant glycogen accumulation in neurons in the form of LBs, which are glycogen-like inclusions with a greater degree of phosphorylation and less branching than normal glycogen. Hence LBs are largely insoluble.

Recent work has demonstrated that neuronal death in LD can be influenced not only by the accumulation of LBs but also by impairment in mechanisms related to protein clearance. For instance, endoplasmic reticulum stress is significantly increased in mouse models of LD [10–13], and molecular interactions between laforin, malin and the chaperone, and proteasome systems [14, 15] have been described. This body of evidence points to a critical dysregulation of cellular protein homeostasis in LD that goes beyond alterations in glycogen metabolism regulation. In this sense, our previous work with mouse models of LD such as Epm2a −/− [16] and Epm2b −/− [17] revealed impairment of autophagy and defects in the ubiquitin-proteasome system [12, 18] which correlated with neurological and behavioral abnormalities. We have also described increased oxidative stress and impaired antioxidant response in LD cellular and mouse models [19], thus suggesting that oxidative stress is a new hallmark in the pathogenesis of LD [20]. These results are in agreement with the hypothesis that defects in autophagy and protein clearance are related with an increase in reactive oxygen species (ROS) production and consequent oxidative stress [21–24]. It is feasible that alterations in protein homeostasis and oxidative stress could underlie the appearance of reactive astrogliosis observed in different brain areas of malin knock-out (Epm2b −/−) mice [25, 26]. Microglia activation and neuroinflammatory responses are currently observed in most neurodegenerative diseases and they are determinants in the pathogenesis of these processes [27–30]. However, very little is known about inflammatory responses and most particularly about gene modulation of cytokines and mediators of the immune response along with disease progression in LD. The present study is focused on possible inflammatory responses in the brain of laforin knock-out (Epm2a −/−) and malin knock-out (Epm2b −/−) mice to learn whether inflammation plays a role in the pathogenesis of LD, in the same way that inflammation has functional implications in the pathogenesis of neurodegenerative diseases with abnormal protein aggregates.

Material and Methods

Ethical Statement, Animal Care, Mice, and Husbandry

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Consejo Superior de Investigaciones Cientificas (CSIC, Spain). All mouse procedures were approved by the animal committee of the Instituto de Biomedicina de Valencia-CSIC [Permit Number: INTRA12 (IBV-4)]. All efforts were made to minimize animal suffering. To eliminate the effect of differences in the genetic background of the animals, we backcrossed Epm2a −/− and Epm2b −/− mice (with a mixed background 129sv:C57BL/6) as previously described [16, 17, 31] with control C57BL/6JRccHsd mice obtained from Harlan laboratories (Barcelona, Spain) ten times to obtain homozygous Epm2a −/− and Epm2b −/− in a pure background. Mice were maintained in the IBV-CSIC facility on a 12/12 light/dark cycle under constant temperature (23 °C) with food and water provided ad libitum.

Male mice of 16 days, 3 months, and 12 months of age were sacrificed by cervical dislocation. Brain was recovered and the two hemispheres separated, one conserved at −80 °C for processing for Western blot and qPCR analyses, and the other fixed in 4 % paraformaldehyde in phosphate buffer saline (PBS) for immunohistochemical analyses.

Immunohistochemical Analyses

Dehydrated tissues from at least three independent mice per group were embedded in paraffin and sectioned at 4 μm. Some sections were stained with periodic acid Schiff (PAS). Other sections were de-waxed, rehydrated, and warmed at 95 °C for 30 min in 10 mM citrate buffer for antigen retrieval. Sections were blocked in blocking buffer (1 % bovine serum albumin; 5 % fetal bovine serum in PBS) and incubated overnight at 4 °C with the appropriate primary antibody diluted in blocking buffer: anti-GFAP (diluted 1/300; Sigma ref. G3893, Saint Louis, MO, USA), Iba-1 (diluted 1/200; Wako ref. 019-19741, Richmond, VA, USA) and anti-Cox2 (diluted 1/250; Cayman Chemicals ref. 160107, Ann Arbor, MI, USA). After three washes of 10 min in PBS, sections were incubated for 1 h at room temperature with the corresponding biotin-conjugated anti-rabbit or anti-mouse secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA) diluted in blocking buffer, washed three times with PBS for 5 min and visualized with the Avidin–Biotin Complex (ABC) (Vectastain Elite ABC kit, Vector Laboratories, Burlingame, CA, USA) using diaminobenzidine as chromogenic substrate for peroxidase (Peroxidase substrate kit DAB; Vector Laboratories, Burlingame, CA, USA). Sections were slightly counterstained with hematoxylin (Sigma, Madrid, Spain), dehydrated, and mounted in DPX (Merck, Germany). Images were acquired by light microscopy (Leica DM RXA2, Leica Microsystems, Wetzlar, Germany) and analyzed with Image J software (NIH, Bethesda, MD, USA). The intensity of the signal was quantified using the Image J software in an area of 0.17 mm2 located between the dentate gyrus and the CA1 region of the hippocampus, and expressed as the number of pixels/mm2. The significance level (Student’s t test) was set at *p < 0.05, **p < 0.01, and ***p < 0.001.

RNA Purification

Total RNA from the posterior part of the neocortex and hippocampus of WT, Epm2a −/−, and Epm2b −/− animals aged 16 days, 3 months, and 12 months (n = 7 per group), respectively, was isolated with the Rneasy Lipid Tissue Mini Kit (Qiagen® GmbH, Hilden, Germany) following the manufacturer’s protocol. RNA concentration of each sample was measured using a NanoDrop™ Spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). RNA integrity number (RIN) was tested using the Agilent 2100 BioAnalyzer (Agilent Technologies, Palo Alto, CA, USA).

TaqMan qRT-PCR

Complementary DNA (cDNA) was prepared using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) following the protocol provided by the supplier. Parallel reactions for each RNA sample were run in the absence of MultiScribe Reverse Transcriptase to assess the lack of contamination of genomic DNA. TaqMan qRT-PCR assays were performed in duplicate for each gene on cDNA samples in 384-well optical plates using an ABI Prism 7900 Sequence Detection system (Applied Biosystems, Life Technologies, Waltham, MA, USA). For each 10 μL TaqMan reaction, 4.5 μL cDNA was mixed with 0.5 μL 20× TaqMan Gene Expression Assays and 5 μL of 2× TaqMan Universal PCR Master Mix (Applied Biosystems). The identification numbers and names of TaqMan probes are shown in Table 1. The mean values of the three house-keeping genes, hypoxanthine-guanine phosphoribosyltranferase (Hprt), alanyl-transfer RNA synthase (Aars), and X-prolyl aminopeptidase (aminopeptidaseP) 1 (Xpnpep1), were used as internal controls for normalization. The reactions were carried out using the following parameters: 50 °C for 2 min, 95 °C for 10 min, and 40 cycles of 95 °C for 15 sec and 60 °C for 1 min. Finally, all TaqMan PCR data were captured using the Sequence Detection Software (SDS version 2.2.2, Applied Biosystems). Samples were analyzed with the double-delta cycle threshold (ΔΔCT) method. Results were analyzed with two-way ANOVA followed by Tukey’s post hoc or Student’s t test when required. The significance level was set at *p < 0.05, **p < 0.01, and ***p < 0.001.

Table 1 TaqMan probes used for the study of expression of cytokines and mediators of the immune response in mouse, including probes used for normalization (Hprt, Aars, and Xpnpep1)
Full size table

Western Blotting

Mouse brain homogenates were lysed in RIPA buffer (50 mM Tris-HCl, pH 7.0; 150 mM NaCl, 1 % Nonidet P-40; 0.5 % Na-deoxycholate; 0.1 % SDS) supplemented with protease and phosphatase inhibitors (Roche, Germany). After centrifugation at 14,000 g for 20 min at 4 °C (Ultracentrifuge Beckman with 70Ti rotor), supernatants were quantified with BCA reagent (Pierce, Waltham, MA, USA). Protein samples were mixed with loading sample buffer and heated at 95 °C for 5 min. Twenty μg of proteins was separated by electrophoresis in SDS-PAGE gels and transferred to nitrocellulose membranes (200 mA per membrane, 80 minutes). Nonspecific binding was blocked by incubation in 5 % non-fat milk in TBS containing 0.2 % Tween for 1 hour at room temperature. The membranes were incubated at 4 °C overnight with primary antibodies: anti-IL1β (diluted 1:1000, Abcam, Cambridge, UK), anti-IL6 (diluted 1:1000; Abcam, Cambridge, UK), anti-TNFα (diluted 1:1000, Abcam, Cambridge, UK) and anti-β-actin (diluted 1:30,000; Sigma-Aldrich, St. Louis, MO, USA). Membranes were washed with TBS-T and incubated for 1 h at room temperature with the appropriate horseradish peroxidase conjugated secondary antibody (1:2,000; Dako, Glostrup, Denmark). Immune complexes were revealed by incubating the membranes with chemiluminescence reagent (electrochemiluminescence; Amersham, GE Healthcare, Buckinghamshire, UK). Densitometries were carried out with Totallab software (TL100 v.2006b), and values were normalized using β-actin levels. Normalized values were expressed as fold change from values obtained in WT aged 16 days samples.

Results

Lafora Inclusions in the Telencephalon in Epm2a −/− and Epm2b −/− Mice

A few PAS-positive round or elongated inclusions in the neuropil and more rarely in the neuronal cytoplasm were first seen in transgenic mice aged 3 months. The number and size of the inclusions was higher in Epm2a −/− and Epm2b −/− mice aged 12 months as detailed elsewhere [16, 17]. Examples of these deposits are shown in Fig 1.

Fig. 1
figure 1

Periodic acid Schiff (PAS)-staining of representative sections of the hippocampus of wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively. a–i: scale bar in I = 150 μm; C1, F1, I1, scale bar in C1 = 15 μm. A few PAS-positive inclusions are seen in Epm2a −/− and Epm2b −/− at the age of 3 months. The number and size of inclusions augments at the age of 12 months

Full size image

Astrocyte and Microglial Responses in Epm2a −/− and Epm2b −/− Mice

We analyzed by immunohistochemistry the levels of GFAP (glial fibrillary acidic protein; a marker of reactive astrocytes) and Iba1 (ionized calcium-binding adapter molecule 1; a marker of reactive microglia) in selected brain areas from wild-type (WT), Epm2a −/−, and Epm2b −/− mice aged 16 days, 3 months, and 12 months, respectively. We observed that GFAP-immunoreactive astrocytes increased in number in the hippocampus (Fig. 2) and cerebral cortex (not shown) of Epm2a −/− and Epm2b −/− mice in comparison to WT littermates, at 3 and 12 months of age. No significant differences were found between Epm2a −/− and Epm2b −/− mice. Iba1-immunoreactive microglia also increased in number in Epm2a −/− and Epm2b −/− mice when compared to WT littermates at the age of 3 and 12 months (Fig. 3). In addition, we found that the number of reactive microglial cells was greater in Epm2b −/− when compared to Epm2a −/− mice (Fig. 3).

Fig. 2
figure 2

Glial fibrillary acidic protein (GFAP) immunohistochemistry of the hippocampus of wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively. Paraffin sections slightly counterstained with hematoxylin. a–i, scale bar in I = 150 μm. The intensity of the signal was expressed as number of pixels/mm2. Data are the mean values from at least three independent mice; vertical bars indicate standard deviation. Marked increase in the number of astrocytes is observed in the CA1 region of the hippocampus in Epm2a −/− and Epm2b −/− mice when compared with WT at the age of 3 months and 12 months. Significant differences with respect to the corresponding WT are evaluated as *p < 0.05, **p < 0.01, ***p < 0.001 compared with age-matched WT. (Student’s t test). No significant differences were observed between Epm2a −/− and Epm2b −/− mice in any age group

Full size image
Fig. 3
figure 3

Iba1 immunohistochemistry of the hippocampus of wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively. Paraffin sections slightly stained with hematoxylin. a–i, scale bar in I = 170 μm. A1, E1, and I1, scale bar in A1 = 15 μm. The intensity of the signal was expressed as the number of pixels/mm2. Data are the mean values from at least three independent mice; vertical bars indicate standard deviation. Marked increase in the number of microglia is observed in the CA1 region of the hippocampus in Epm2a −/− and Epm2b −/− mice when compared with WT at the age of 3 months and 12 months. Significant differences with respect to the corresponding WT are evaluated as *p < 0.05, **p < 0.01, ***p < 0.001 compared with age-matched WT  ▫p < 0.05 comparedwith Epm2a -/- (Student’s t test)

Full size image

Gene Expression of Cytokines and Mediators of the Inflammatory Response with Aging in WT, Epm2a −/− and Epm2b −/− Mice

Since reactive astrocytes and microglia are mediators of the innate inflammatory response, we assessed the age-dependent expression of a panel of cytokines and mediators of the inflammatory response in wild-type, Epm2a −/−, and Epm2b −/− mice. First, we analyzed the regular age-dependent changes in the expression of these genes in wild-type mice: we found that C3ar1, C4b, Csf3r, Ccl6, CxCl10, Il6, Il6st, Tnfrsf1a, Il10ra, and Il10rb messenger RNA (mRNA) expression levels significantly increased and C1ql1 levels significantly decreased in WT mice of 3 months of age when compared with WT mice aged 16 days. The differences in the expression levels of these genes further increased in WT mice aged 12 months when compared with mice aged 16 days; in addition, Ccl3, Ccl4, Tlr4, and Tnfα mRNA expression was higher in WT mice aged 12 months when compared with mice aged 16 days. However, Tgf-β1 decreased with age. Increased expression with age was supported to a greater extent by the fact that C4b, Ccl3, Ccl4, Ccl6, CxCl10, Il6, and Tnfα were significantly increased in WT mice aged 12 months when compared with WT mice 3 months old (Table 2).

Table 2 mRNA expression of selected cytokine-related genes involved in the inflammatory response in wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively
Full size table

In the case of Epm2a −/− mice, C3ar1, Csf3r, Ccl4, Il1b, Il6st, Il10ra, and Il10rb mRNAs were increased and C1ql1 mRNA was decreased in mice aged 3 months when compared with mice aged 16 days. Moreover, C3ar1, C4b, Csf1r, Csf3r, Tlr4, Ccl4, Ccl6, CxCl10, Il1b, Il6, Tnfα, Il10ra, and Il10rb mRNAs were significantly increased in Epm2a −/− aged 12 months when compared with Epm2a −/− aged 16 days. C1ql1 mRNA also significantly decreased in Epm2a −/− mice aged 12 months. Finally, nine genes, C3ar1, C4b, Csfr1, Csf3r, Ccl6, CxCl10, Il1b, Il6, and Tnfα, were significantly upregulated in Epm2a −/− aged 12 months when compared with Epm2a −/− aged 3 months (Table 2).

Regarding modifications in gene expression with aging in Epm2b −/− mice, C3ar1, Ccl6, Il6, Il6st, and Il10rb mRNA expression was increased and C1ql1 mRNA levels were downregulated in mice aged 3 months when compared with 16-day-old mice. C3ar1, C4b, Csf3r, Tlr4, Ccl3, Ccl4, Ccl6, CxCl10, Il1b, Il6, Il6st, Tnfα, Tnfrsf1a, Il10, Il10ra, and Il10rb were significantly increased in Epm2b −/− mice aged 12 months when compared with Epm2b −/− mice aged 16 days. In contrast, a significant reduction in C1ql1 mRNA levels was observed in Epm2b −/− mice aged 12 months when compared with younger animals. Differences with age were greater in Epm2b −/− when compared with Epm2a −/− mice, as thirteen genes instead of nine genes were significantly upregulated in Epm2b −/− mice aged 12 months when compared with mice aged 3 months, including C3ar1, C4b, Tlr4, Tlr7, Ccl3, Ccl4, Ccl6, CxCl10, Il1b, Il6, Tnfα, Il10ra, and Il10rb (Table 2).

Differential Gene Expression of Cytokines and Mediators of the Inflammatory Response in Epm2a −/− and Epm2b −/− Mice Compared with WT Mice

Then, we analyzed the differences in gene expression at the different ages between wild-type, Epm2a −/−, and Epm2b −/− mice. Few differences were seen in Epm2a −/− (laforin knock-out) and Epm2b −/− (malin knock-out) mice aged 16 days when compared with age-matched WT littermates; we found a small increase in the expression of Il6 and Tnfα in Epm2a −/− mice and a small increase in the expression of C1qtnf7, Csf3r, and Il6 in Epm2b −/− mice at this age (Table 2). At 3 months of age we found an increase in the expression of Ccl4, CxCl10, and Il6st in Epm2a −/− mice, and of CxCCl10 and Tnfα in Epm2b −/− mice (Table 2). In contrast, a significant reduction of Ccl4 and Il10ra occurred in Epm2b −/− when compared with Epm2a −/− mice at this age. However, a dramatic gene upregulation was found in Epm2b −/− mice at the age of 12 months when compared with the limited change in Epm2a −/− mice at a similar age. Whereas only C3ar1, Csf3r, and CxCl10 mRNA expression was increased in Epm2a −/− in comparison to WT, the expression of C3ar1, C4b, Ccl4, Ccl6, CxCl10, Il1b, Il6, Tnfα, and Il10ra was upregulated in Epm2b −/− aged 12 months when compared with age-matched WT (Table 2). All these results are compatible with the presence of inflammation in the brain of LD mice at 3 months of age becoming aggravated as the mice aged, with the phenotype more severe in the case of Epm2b −/− mice. In fact, C3ar1, Ccl4, Ccl6, CxCl10, Il1b, Tnfα, and Il10ra mRNA expression was significantly upregulated in Emp2b −/− mice aged 12 months when compared with Epm2a −/− mice of the same age (Table 2).

Protein Expression of Selected Cytokines and Mediators

In order to confirm the inflammatory process present in the brain of Epm2a −/− and Epm2b −/− mice, IL1-β, IL6, and TNFα protein levels were analyzed by Western blotting in brain extracts from wild-type, Epm2a −/−, and Epm2b −/− mice aged 16 days, 3 months, and 12 months, respectively. As shown in Fig. 4, at 16 days of age Epm2b −/− mice showed an increase in the levels of IL1-β and IL6 in comparison to WT mice. No differences were observed in the case of Epm2a −/− mice. At 3 months of age, we found in Epm2b −/− mice a tendency to higher levels of IL1-β and IL6 and a significant increase in the levels of TNFα in comparison to WT of the same age. However, Epm2a −/− mice showed similar protein levels of IL1-β and IL6 and reduced levels of TNFα at this age in comparison to WT. At 12 months of age, Epm2b −/− mice showed higher levels of IL1-β and IL6 but similar levels of TNFα in comparison to WT of the same age. In contrast, Epm2a −/− mice showed levels of IL1-β, IL6, and TNFα similar to WT. These data confirm that Epm2b −/− mice present a more severe inflammatory pattern than Epm2a −/− mice.

Fig. 4
figure 4

Western blot of IL1-β, IL6, and TNFα in wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively. Data show an increase in IL1-β and IL6 protein levels in Epm2b −/− aged 16 days and 12 months when compared with age-matched WT. A significant reduction in TNFα protein levels occurred in Epm2a −/− and an increase in Epm2b −/− aged 3 months when compared to WT. β-actin is used as a control of protein loading. Data are presented as the mean ± SEM. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with age-matched WT. ▫p < 0.05, ▫▫p < 0.01, ▫▫▫p < 0.001 compared with Epm2a −/−. a p < 0.05, aa p < 0.01, aaa p < 0.001 compared with animals aged 16 days. b p < 0.05, bb p < 0.01, bbb p < 0.001 compared with animals aged 3 months

Full size image

In addition, we analyzed the levels of cyclooxygenase 2 (Cox2), a target gene of inflammatory mediators. An immunohistochemical analysis of the hippocampus of WT, Epm2a −/−, and Epm2b −/− mice of different ages showed an increase in the levels of Cox2 in the case of Epm2a −/− and Epm2b −/− mice at 3 months of age, which dramatically increased at 12 months of age (Fig. 5), with the staining at this age more intense in the case of Epm2b −/− in comparison to Epm2a −/− mice.

Fig. 5
figure 5

Cox2 immunohistochemistry of the hippocampus of wild-type (WT), Epm2a −/− (laforin knock-out), and Epm2b −/− (malin knock-out) mice aged 16 days, 3 months, and 12 months, respectively. Paraffin sections slightly stained with hematoxylin. a–i, scale bar in I = 170 μm, I1, scale bar = 15 μm. The intensity of the signal was expressed as the number of pixels/mm2. Data are the mean values from at least three independent mice; vertical bars indicate standard deviation. Marked increase in the levels of Cox2 is observed in the CA1 region of the hippocampus in Epm2a −/− and Epm2b −/− mice when compared with WT at the age of 3 and 12 months. Significant differences with respect to the corresponding WT are evaluated as *p < 0.05, **p < 0.01, ***p < 0.001 compared with age-matched WT. ▫p < 0.05 compared with Epm2a −/−; a p < 0.05, aa p < 0.01 compared with animals aged 3 months (Student’s t test)

Full size image

Discussion

Laforin knock-out (Epm2a −/−) and malin knock-out (Epm2b −/−) mice used in the present study [17, 31] start to develop LBs in the cerebral cortex, hippocampus, basal ganglia, thalamus, cerebellum, cerebellar nuclei, and brain stem at about the first month of age, and this increases in number and size with disease progression. LBs are very abundant in these regions at advanced stages of the disease [17, 31, 32]. Deposits are also found in the choroid plexus but the olfactory bulb is largely spared. Although both mouse lines have clinical symptoms similar to those seen in human LD, Epm2b −/− mice show some differences in comparison to Epm2a −/− mice such as reduced accumulated motor surface, rearing and stereotyped activity, as well as lower episodic memory deficits and non-spatial memory skills. In addition, although both mouse lines have tonic-clonic seizures, these occur in parallel with spike-wave, poly-spikes, and poly-spike-wave complexes in Epm2a −/− mice but there are no EEG correlates in Epm2b −/− mice [32]. All these data suggest that the phenotype observed in Epm2b −/− mice is more severe than that presented in Epm2a −/− mice.

In this study, we investigated whether neuroinflammation is a novel hallmark in Lafora disease. With this aim, we analyzed how the expression of a panel of cytokines and mediators of the inflammatory response changes with age in WT, Epm2a −/−, and Epm2b −/− mice. It is known that marked modification of the profile of gene expression of cytokines and molecules linked with the inflammatory response occurs in the cerebral cortex and other brain regions of mice with age. This happens in normal aging and also in transgenic mice including APP/PS1, P301S-MAPT, and S49P-Syracuse human neuroserpin (S49P-Syracuse), and in CJD-induced prionopathy in murine PrP-null mice expressing human PrP, which are models of human β-amyloidopathy reminiscent of Alzheimer disease, familial frontotemporal degeneration-tauopathy (FTLD-tau), neuroserpinopathy, and Creutzfeldt-Jakob disease, respectively [33–36]. This pattern is repeated in our study as C3ar1, C4b, Csf3r, Tlr4, Ccl4, Ccl6, CxCl10, Il1b, Il6, Tnfα, Il10ra, and Il10rb mRNA expression levels increase and C1ql1 levels significantly decrease in WT, Epm2a −/−, and Epm2b −/− mice aged 12 months when compared with mice aged 16 days. At the same time, Ccl3, Il6st, and Tgfb1 are increased only in WT and Epm2a −/− mice whereas Tnfrsf1a and Il10 increased only in Epm2b −/− mice. These findings support the idea that gene expression of cytokines and mediators of the inflammatory response is modulated with age, albeit with genotypic variations.

Focusing on pathologic states, inflammatory responses are increased in Epm2a −/− and Epm2b −/− mice when compared with WT littermates. C3ar1, Csf3r, and CxCl10 are significantly increased in Epm2a −/− mice aged 12 months when compared with age-matched controls. Curiously, inflammatory responses are greater in Epm2b −/− mice when compared with Epm2a −/− mice, as C3ar1, C4b, Ccl4, Ccl6, CxCl10, Il1b, Il6, Tnfα, and Il10ra mRNAs are significantly upregulated in Epm2b −/− aged 12 months. In general terms, IL1-β, IL6, and TNFα protein levels are in accordance with the corresponding mRNA expression levels, with the Epm2b −/− genotype the most dramatically affected. However, a discrepancy exists between Tnfα mRNA and TNFα protein levels at the age of 3 months in Epm2a −/−. It can be suggested that unexpected reduced TNFα protein expression is related to epigenetic regulation. However there is no available information about non-coding RNAs in LD and related transgenic models. An explanation of this punctual finding at the age of 3 months in Epm2a −/− is not known. In agreement with increased expression of cytokines and mediators of the immune response in both LD mouse models, we found increased levels of Cox2, a protein with pro-inflammatory properties, in both Epm2a −/− and Epm2b −/− mice, although we observed higher protein levels in Epm2b −/− mice as the animals aged.

We also present data indicating that the number of GFAP-immunoreactive astrocytes increases in Epm2a −/− and Epm2b −/− mice when compared with WT littermates, although this increase is similar in Epm2a −/− and in Epm2b −/− mice. Therefore, the extent of dysregulated gene expression of cytokines and mediators of the inflammatory response, which is higher in Epm2b −/− mice, cannot be explained by a differing number of reactive astrocytes. However, we also present evidence of an increase in Iba-1-immunoreactive cells (microglia) in Epm2a −/− and Epm2b −/− mice, which is greater in Epm2b −/− mice in comparison to Epm2a −/− mice. This phenotype correlates with the increased gene expression of cytokines and mediators of the inflammatory response present in Epm2b −/− mice, suggesting involvement of reactive microglia in the severity of the inflammatory phenotype of Emp2b −/− mice.

Therefore, and in agreement with previous data [32], the phenotype in Epm2b −/− mice is more severe than in the Epm2a −/− mice. The causes of the more severe phenotype in Epm2b −/− are not known but our present results clearly show a close relationship between stronger microglia-dependent inflammatory responses and the already described impaired motor function, memory, and learning, together with the unpredictable seizures present in these mice. Microglia may have very disparate functions depending on the setting and stimulus [37], but chronic inflammation, involving chemokines, pro-inflammatory cytokines, anti-inflammatory cytokines, and members of the complement system, is usually associated with nerve cell damage in human neurodegenerative diseases with abnormal protein aggregates and in related animals models. The observations presented in this study are particularly relevant since increased microglia and marked increase in mRNA and protein expression of cytokines and mediators of the inflammatory response are described for the first time in LD, a paradigm of polyglucosan disease. Insoluble phosphorylated polyglucosans, endoplasmic reticulum stress, oxidative stress, altered autophagy, and ubiquitin-proteasome activity may all contribute to the inflammatory response. Since inflammatory responses appear at relatively early stages of LD, the administration of anti-inflammatory agents may be considered a possible treatment in order to delay brain damage.