mTOR Hyperactivation in Down Syndrome Hippocampus Appears Early During Development

Abstract

The mammalian target of rapamycin (mTOR) signaling pathway is a key developmental pathway involved in mechanisms underlying cellular aging and neurodegeneration. We hypothesized that its deregulation may occur during early brain development in patients with Down syndrome (DS). The expression patterns and cellular distribution of components of mTOR signaling (phosphorylated S6, phosphorylated S6 kinase, phosphorylated eukaryotic initiation factor 4E binding protein 1, and phosphorylated mTOR) were investigated in developing hippocampi from controls and patients with DS and from adults with DS and Alzheimer disease–associated pathology using immunocytochemistry. In control hippocampi, only phosphorylated S6 was detected prenatally (19–41 gestational weeks); it became undetectable 2 months postnatally. Increased expression of phosphorylated S6, phosphorylated S6 kinase, phosphorylated eukaryotic initiation factor 4E binding protein 1, and phosphorylated mTOR was observed in DS hippocampus compared with controls. Phosphorylated S6 and phosphorylated S6 kinase were detected prenatally and persisted throughout postnatal development. Prominent expression of mTOR components was observed in pyramidal neurons with granulovacuolar degeneration and in neurons containing neurofibrillary tangles in the hippocampi of DS subjects with Alzheimer disease pathology. These findings suggest that a dysregulated mTOR pathway may contribute to both early hippocampal developmental abnormalities and hippocampal functional impairment developing before neurodegeneration. Moreover, the expression patterns of mTOR components in adult DS hippocampus support its association with Alzheimer disease–related histopathologic changes.

Introduction

Increasing evidence supports the idea that neurologic disorders could represent a disorder of aberrant neural development. This “developmental hypothesis” could even be applied to neurodegenerative disorders such as Alzheimer disease (AD) (1). Accordingly, major signaling pathways essential for normal cortical development have been implicated in the mechanisms underlying cellular aging and neurodegeneration (1–3). In particular, attention has been recently focused on the role of mammalian target of rapamycin (mTOR), a ubiquitous serine/threonine kinase that functions in response to a large variety of environmental stimuli and regulates essential physiologic functions, including cell growth, proliferation, protein synthesis, metabolism, and autophagy (4–6). Aberrant hyperactivation of this pathway has been recently shown in different developmental disorders characterized by complex neurobehavioral disabilities (6–8). Moreover, several studies indicate that the mTOR may also critically contribute to accumulation of cell damage, promoting the development of aging-related diseases (5, 6, 9). In particular, an interrelationship between mTOR signaling and AD-like neuropathology has been recently shown in mouse models of AD, pointing to the mTOR pathway as a molecular link between amyloid β accumulation and cognitive dysfunction (10, 11). Accordingly, mTOR inhibitors have been shown to reverse cognitive deficits and to reduce amyloid β levels in mouse models of AD (10, 12). Moreover, mTOR and its downstream signaling kinases have been shown to be upregulated in the brains of patients with AD (13).

Down syndrome (DS), or trisomy 21, is one of the most common genetic causes of cognitive impairment; aged individuals with DS develop progressive AD neuropathology (14–16). The aberrant brain development of patients with DS represents an interesting model of accelerated aging, offering researchers the opportunity to understand the link between development and aging and to clarify many unclear aspects of the pathogenesis of AD, including the deregulation of specific signaling pathways.

Recent studies have indicated that there is developmental regulation of mTOR complex (mTORC) 1 signaling (17), but the expression pattern of its downstream targets during normal human hippocampal development and in the hippocampi of DS patients remains uncharacterized.

In the present study, we investigated the expression and cell-specific distribution of the components of the mTOR pathway during prenatal and postnatal human hippocampal development. In addition, because there is a potential clinical link between mTOR deregulation, neurobehavioral deficits, and AD pathology, we investigated the mTORC1 downstream targets in the developing hippocampus of DS patients before establishment of AD neurodegeneration and in DS cases with advanced AD pathology.

Materials and Methods

Human Cases

The subjects included in this study were selected from the databases of the Department of Neuropathology, Academic Medical Center, University of Amsterdam (Amsterdam, the Netherlands) and the Institute of Neurology, Medical University of Vienna (Vienna, Austria; in the frame of a project approved by the Ethics Committee of the Medical University of Vienna, “Molecular neuropathologic examinations of neurodegeneration-related proteins in Down syndrome,” Ek No. 1316/2012); from the National Institute of Child Health and Human Development Brain and Tissue Bank for Developmental Disorders (University of Maryland, Baltimore, MD); and from the Netherlands Brain Bank. Informed consent was obtained for the use of brain tissue and for access to medical records for research purposes. Tissue was obtained and used in a manner compliant with the Declaration of Helsinki and with the Academic Medical Center Research Code provided by the Medical Ethics Committee of the Academic Medical Center (Amsterdam, the Netherlands). We examined the donated brains of control and DS fetuses (from 9 to 41 gestational weeks [GW]), neonates, and children (Table 1). Fetal brains were preserved after spontaneous or induced abortion with appropriate maternal written consent for brain autopsy. Gestational ages were based on obstetric data, fetal weight, brain weight, and standard fetal anthropometric measurements. We performed a careful histologic and immunohistochemical analysis and evaluation of clinical data (including genetic data, when available). We excluded cases with other chromosomopathies, major CNS malformations, brains with postmortem autolysis, severe hypoxic/ischemic encephalopathy, intraventricular hemorrhages, severe hydrocephalus, and meningitis or ventriculitis. As control cases, we only included specimens displaying normal hippocampal and cortical structures for the corresponding age and without any significant pathology. In addition, we obtained adult brain tissue at autopsy from 11 controls (without evidence of degenerative changes and lacking a clinical history of cognitive impairment), 6 patients with DS (neurofibrillary degeneration [NFD] Stages V and VI), and 6 patients with AD (NFD Stages V and VI) (Table 1). In all cases, pathology was staged according to criteria recommended for NFD (19) and granulovacuolar degeneration (GVD) (18). All autopsies were performed within 24 hours of death.

Tissue Preparation

One or 2 hippocampal tissue paraffin blocks per case were sectioned, stained, and assessed; hippocampal bodies were cut in the frontal plane. Formalin-fixed paraffin-embedded tissue was sectioned at 6 μm and mounted on precoated glass slides (Star Frost; Waldemar Knittel GmbH, Braunschweig, Germany). Sections were stained with hematoxylin and eosin, Luxol fast blue, and Nissl stains and were processed for immunocytochemical staining of antigens listed in Table 2.

Immunocytochemistry

Single-label immunocytochemistry was developed using the Powervision kit (Immunologic, Duiven, the Netherlands). 3,3-Diaminobenzidine (Sigma, St Louis, MO) was used as chromogen. Sections were counterstained with hematoxylin.

For double-label immunofluorescent staining, sections were incubated with the primary antibodies overnight at 4°C, then incubated for 2 hours at room temperature with Alexa Fluor 568–conjugated anti-rabbit IgG and Alexa Fluor 488 anti-mouse IgG (1:100; Molecular Probes, Leiden, the Netherlands). Sections were then analyzed using a laser scanning confocal microscope (Leica TCS Sp2, Wetzlar, Germany).

For double labeling with antibodies directed against serine/threonine-specific casein kinase I δ (CKIδ) and lysosome-associated membrane protein-2, sections were stained with the first primary antibody, followed by Brightvision poly–alkaline phosphatase (AP) anti-mouse antibody (Immunologic), for 30 minutes at room temperature and washed with PBS. Sections were washed with Tris-HCl buffer (0.1 mol/L, pH 8.2) to adjust pH. Alkaline phosphatase activity was visualized with the AP substrate kit I Vector Red (SK-5100; Vector Laboratories Inc, Burlingame, CA). To remove the first primary antibody, we incubated sections at 121°C in citrate buffer (10 mmol/L NaCi, pH 6.0) for 10 minutes. Incubation with the second primary antibody was performed overnight at 4°C. Sections with primary antibody other than rabbit were incubated with postantibody blocking from the Brightvision+ system (containing rabbit α-mouse IgG; Immunologic) then with Brightvision poly-AP anti-mouse antibody (Immunologic). Alkaline phosphatase activity was visualized with the AP substrate kit III Vector Blue (SK-5300; Vector Laboratories). Sections incubated without or with the primary antibodies, followed by heating treatment, were essentially blank.

Western Blot Analysis

For immunoblot analysis, we used frozen brain specimens from control hippocampi (n = 3 men; mean ± SE age, 58 ± 4.2 years) and AD-DS hippocampi (n = 3 men; mean ± SE age, 59 ± 4.5 years). The frozen specimens were homogenized in lysis buffer containing 10 mmol/L Tris (pH 8.0), 150 mmol/L NaCl, 10% glycerol, 1% NP-40, 0.4 mg/mL Na-orthovanadate, 5 mmol/L ethylenediaminetetraacetic acid (pH 8.0), 5 mmol/L NaF, and protease inhibitors (cocktail tablets; Roche Diagnostics, Mannheim, Germany). Protein content was determined using the bicinchoninic acid method. For electrophoresis, equal amounts of proteins (50 μg per lane) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (12% acrylamide). Separated proteins were transferred to nitrocellulose paper by electroblotting for 1 hour and 30 minutes (Transblot SD; BioRad, Hercules, CA). After blots were blocked for 1 hour in TBST (20 mmol/L Tris, 150 mmol/L NaCl, and 1% Tween, pH 7.5)/5% nonfat dry milk, they were incubated overnight at 4°C with rabbit anti–phosphorylated S6 (pS6) (1:1000) or S6 (1:1000, 5G10; Cell Signaling Technology, Danvers, MA). After several washes in TBST, the membranes were incubated in TBST/5% nonfat dry milk containing goat anti-rabbit or rabbit anti-mouse antibodies coupled with horseradish peroxidase (1:2500; Dako, Glostrup, Denmark) for 1 hour. After washes in TBST, immunoreactivity (IR) was visualized using ECL PLUS Western blot detection reagent (GE Healthcare Europe, Diegen, Belgium). For quantification of blots, band intensities were measured densitometrically by Scion Image for Windows (beta 4.02) image analysis software. The ratio of the integrated band density of the protein of interest to the integrated band density of the reference protein was used to normalize band intensities.

Evaluation of Immunostaining

All labeled tissue sections were evaluated by 2 independent observers (who were blind to clinical data) for the presence or absence of various histopathologic parameters and specific IR to the different antigens. The intensity of staining was evaluated as previously described (20, 21); a semiquantitative scale ranging from 0 to 3 (0, negative staining; 1, weak staining; 2, moderate staining; 3, strong staining) was applied to labeled regions (ventricular zone [VZ]/subventricular zone [SVZ] and pyramidal layer [PL]) (Table 3). Different areas of the PL (CA1–CA4) were examined. Because no major differences in expression patterns were observed for the different antigens within the PL, the intensity score represents the predominant cell staining intensity found for each case in the VZ/SVZ and PL. Immunoreactivity was not detected in the dentate gyrus. The relative numbers of labeled cells (frequency score) within the PL were estimated using a semiquantitative scale as follows: score 1 (rare immunolabeling in <10% of cells); score 2 (sparse immunolabeling in 11%–50% of cells); score 3 (high immunolabeling in >50% of cells). The overall score (total IR score) represents the product of the 2 scores (frequency scores × intensity scores = IR score [20–22]). Total labeling scores in CA1 are shown in Figure 1.

Results

mTORC1 Signaling in the Developing Control Hippocampus

The expression patterns of the components of mTOR signaling (pS6, phosphorylated S6 kinase [p70S6K], phosphorylated eukaryotic initiation factor 4E binding protein 1 [p4E-BP1], and phosphorylated mTOR [pmTOR]) in hippocampi were investigated immunocytochemically at different prenatal ages (ranging from 9 to 41 GW) and postnatal ages (<1 year [1 day to 8 months] and 1 to 15 years) (Tables 1, 3). Phosphorylated 70S6K (Thr389 and Thr229), p4E-BP1 (Thr37/Thr46), and pmTOR (Ser2448 and Ser2481) were not detected through hippocampal development (Figs. 2A, G, 3D, E, 4C). Only pS6 was detected prenatally between 19 and 41 GW; by 2 months postnatally, it was no longer detected (Figs. 1A, 5A, D, H, 6C; Table 3; Figure, Supplemental Digital Content 1, Supplementary Data). Phosphorylated S6 IR was not detected in adult control hippocampus (Fig. 6E). We used 2 antibodies directed against different phosphorylation sites of S6 (Ser235/Ser236 and Ser240/Ser244) (Table 2); similar IR patterns were observed using both antibodies, with few positive cells within the PL but not in the VZ. Ser235/Ser236 is shown in Figure 5.

mTORC1 Signaling Pathway in DS Hippocampi

In DS hippocampi, there were higher levels of expression of pS6, p70S6K, p4E-BP1, and pmTOR versus those in controls (Figs. 1–6; Table 3). Phosphorylated S6 (Ser235/Ser236 and Ser240/Ser244) and p70S6K (Thr389 and Thr229) were detected prenatally and persisted throughout postnatal development (Figs. 1A, B).

Phosphorylated S6 expression was detected at 14 GW in both VZ and PL (Figs. 5B, C; Table 3) and gradually increased in the PL between 14 and 41 GW before decreasing postnatally (Figs. 1A, 5F, G, I, 6A, B, D). Phosphorylated S6 was also was observed in astroglial cells in DS hippocampi (SVZ; between 23 GW and 8 months). Phosphorylated S6–immunoreactive neurons were still detected postnatally (Fig. 6; Figure, Supplemental Digital Content 1, Supplementary Data) even in a 15-year-old DS patient (Fig. 6D); pS6-positive neurons with vacuolar changes (GVD) were also occasionally seen (Fig. 6D, inset b). We did not detect differences in the expression patterns of the non-pS6 protein in DS and control hippocampi (both prenatally and postnatally) (Figure, Supplemental Digital Content 2, Supplementary Data).

Phosphorylated 70S6K (Thr389 and Thr229) IR was also observed prenatally and increased gradually during postnatal development (Fig. 1B; Table 3). Phosphorylated 70S6K was only detected in neurons within the PL (Figs. 2B, D–F; Table 3). Phosphorylated 4E-BP1 (Thr37/Thr46) and pmTOR (Ser2448 and Ser2481) were only detected postnatally in DS hippocampi in few neurons within the PL (Figs. 1C, D, 3B, 4A; Table 3). Phosphorylated 70S6K, p4E-BP1, and pmTOR-positive neurons with GVD were also occasionally seen in a 15-year-old DS patient (Figs. 2F, 3C, 4B).

mTORC1 Signaling Pathway in DS Hippocampi With AD Pathology

Prominent expression of all 4 markers (pS6, p70S6K, p4E-BP1, and pmTOR) was detected in patients with DS hippocampus with AD pathology (DS-AD) (Figs. 1A–D; Table 3). Phosphorylated S6, p70S6K, p4E-BP1, and pmTOR IR was observed in hippocampal pyramidal neurons with GVD; pS6, p70S6K, and p4E-BP1 were also localized in neurons containing neurofibrillary tangles (Figs. 2H–J, 3F–H, 4D, E, 6F–I). Double-label experiments showed colocalization of pS6 with phosphorylated tau and phosphorylated transactive response DNA binding protein, which is known to stain GVD (18) (Fig. 6G), CKIδ (Fig. 6H), and lysosome-associated membrane protein-2 (data not shown).

Western blot analysis could only be performed in adult DS-AD cases for which frozen material was available. Expression of pS6 was increased in patients with AD pathology compared with that in controls (Figs. 7A, B).

Discussion

This study provides the first description of the expression patterns and cellular distribution of the components of the mTORC1 pathway in human hippocampi during prenatal and early postnatal development in controls and patients with DS.

Increasing evidence supports the role of mTOR in the pathogenesis of aging-related diseases, including AD (5, 6, 9). In particular, several experimental studies point to mTOR signaling as a molecular link between amyloid β accumulation and cognitive dysfunction (10–12). However, it is not clear whether (or to what extent) aberrant hyperactivation of this pathway may occur in DS patients even before the establishment of AD neurodegeneration.

We observed the developmental regulation of pS6, which was detected prenatally between 19 and 41 GW and was undetected by 2 months postnatally. Developmental regulation of mTORC1 signaling has been recently reported in rodent hippocampi with a transient pS6 upregulation during the first 2 postnatal weeks, corresponding to a period of active synaptogenesis (17). Similar developmental changes were also observed for p70S6K, p4E-BP1, and pmTOR on Western blot analysis in rodent hippocampi (17). In our study, only pS6 was observed (at least to a level detectable by immunocytochemistry) in control human hippocampi.

We need to take into account the fact that S6 can be phosphorylated by different kinases at different phosphorylation sites. In particular, S6 can be phosphorylated at Ser235/Ser236 by p70S6K, but also by p90 ribosomal S6 kinase (23). However, evaluation of pS6 phosphorylated at Ser240/Ser244, which is specifically phosphorylated by p70S6K, displayed IR patterns similar to those of pS6 phosphorylated at Ser235/Ser236. Phosphorylated S6 was detected in neuronal cells within the PL, as was reported in rats (17).

The increase in pS6 corresponds to the developmental upregulation of upstream activators of mTOR, including glutamate receptors such as metabotropic glutamate receptor 5 (mGluR5) (24–26). Accordingly, activation of the mTOR pathway has been reported after stimulation of mGluR5 (27–31).

Recently, Talos et al (17) reported time-dependent activation of downstream components (such as pS6) of the mTOR pathway after hypoxia-induced seizure onset and lasting up to 24 hours in the immature rat brain. In view of the potential effect of the regulation of mTOR by hypoxia, we excluded from our present cohort those cases with severe hypoxic/ischemic encephalopathy and seizures. In a recent study on human neonatal hippocampus after perinatal asphyxia, only few pS6-positive neuronal cells were occasionally observed within the hippocampus, even in patients with seizures (32). Moreover, hypoxia itself has been shown to suppress, rather than activate, mTOR signaling (33, 34).

Down syndrome patients develop dementia with clinical and neuropathologic features similar, although not identical, to those of adults with AD without DS (35–38); this offers researchers the opportunity to understand the link between aberrant neural development and pathologic premature aging (1–3). The present study shows upregulation of mTOR downstream components (pS6, p70S6K, p4E-BP1, and pmTOR) in DS hippocampi compared with controls. In particular, pS6 and p70S6K were detected prenatally and persisted throughout postnatal development (with higher pS6 levels around birth), whereas p4E-BP1 and pmTOR were detected postnatally in DS hippocampi. The slightly different timing of expression observed for pS6 compared with other mTOR components, including p70S6K, may reflect phosphorylation by different kinases at different sites (23, 39, 40). In addition to phosphorylation at Thr389, phosphorylation at Thr229 by phosphorylated 3-phosphoinositide dependent protein kinase-1 (pPDK1) may critically regulate the activity of p70S6K (41); phosphorylation of mTOR at Ser2448 has been shown to be associated with both mTORC1 and mTORC2 (42). However, we did not observe major differences in the expression patterns of pS6, mTOR, and p70S6K antibodies used in this study. Moreover, there may be negative feedback loop regulation mechanisms in the control of mTOR activity that involve crosstalk between p70S6K and phosphoinositide-3-kinase/Akt signaling (43). Therefore, we acknowledge limitations to the interpretation of the present immunocytochemical results. Frozen hippocampal material at different developmental ages was not available for additional Western blot analysis, and the functional relevance of different phosphorylation sites remains unclear and deserves further investigations that are beyond the scope of this study.

All 4 mTOR components evaluated (pS6, p70S6K, p4E-BP1, and pmTOR) showed neuronal expression within the PL. Only pS6 was additionally transiently detected in astroglial cells in the VZ/SVZ. Accordingly, expression of pS6 has been previously reported in astrocytes (particularly reactive astrocytes), and recent data support the role of mTOR in the regulation of glial functions (44, 45). The mechanisms underlying the activation and regulation of mTOR signaling in neuronal and glial cells in DS hippocampi remain to be determined. As discussed previously, we cannot exclude the role of upstream activators of mTOR, such as mGluR5. Accordingly, previous studies reported upregulation of mGluR5 in the cortex of DS patients (46).

This study also reveals prominent neuronal expression of all 4 mTOR downstream components in DS-AD patients. This is in agreement with a previous study showing increased levels of pmTOR (Ser2481) and p4E-BP1 (Thr70 and Ser65) in the medial temporal cortex of AD patients; there was also a positive correlation with total tau and phosphorylated tau and expression in neuronal cells, including some tangle-like neurons for p4E-BP1 (13). In our study, expression of pS6, p70S6K, and p4E-BP1 was also detected in neurons containing neurofibrillary tangles within DS-AD. Recent functional studies indicate that tau-related kinases are targets of mTOR, and the use of inhibitors of mTOR pathways has been suggested as a new strategy for reducing or preventing tau hyperphosphorylation (47). Recently, enhanced levels of phosphorylated tau protein have also been reported within the dysplastic cortex of patients with hemimegalencephaly (48) and focal cortical dysplasia (49)—malformations of cortical development that are characterized by mTOR hyperactivation.

In addition, other studies indicate that autophagosome lysosomal degradation is impaired in AD and that the mTOR pathway, being critically involved in the regulation of autophagy, could represent an effective therapeutic target for AD (9). In the hippocampi of DS-AD patients, we observed a consistent expression of pS6, p70S6K, p4E-BP1, and pmTOR in hippocampal pyramidal neurons with GVD. Strong neuronal pS6 IR has been recently reported in structures that corresponded morphologically to GVD in AD hippocampi (50). Granulovacuolar degeneration is a marker of aging and neurodegeneration (51). Neurons with GVD are often encountered in the brains of patients with AD, and the recently described GVD stages are associated with the stages of AD pathology, supporting the possible role of GVD in the pathogenesis of AD (52). The role of GVD in DS patients has not been systematically examined, and its significance as an early disease marker remains a matter of discussion (53). Recently, a defect in lysosome fusion efficiency, combined with mTOR-mediated suppression of phagophore formation, has been suggested to play a role in the formation of GVD bodies (52). In our study, GVD lesions were observed in all adult DS-AD patients and displayed positivity for the different markers of mTOR activity. Interestingly, neurons with GVD and IR to these markers were also occasionally seen in DS (a 15-year-old patient) before end-stage AD pathology and dementia were established. Thus, future investigations in a large patient cohort (including also postnatal ages ranging between 10 and 40 years) would be necessary to clarify the role of GVD and its relationship with mTOR pathway activation in DS.

In conclusion, our study provides evidence of mTOR pathway activation in DS hippocampi. Because DS can serve as a model of AD-related pathology, the alteration detected early during brain development supports the notion that mTOR is an important pathway in AD and that neurodegeneration-related changes are initiated in the brain early during development.

References

Supporting Information

Supplementary Data

Author notes