Calpain I Activation Causes GLUT3 Proteolysis and Downregulation of O-GlcNAcylation in Alzheimer’s Disease Brain

Abstract

Impairment of cerebral glucose uptake/metabolism in individuals with Alzheimer’s disease (AD) is believed to lead to downregulation of protein O-GlcNAcylation, which contributes to tau pathogenesis through tau hyperphosphorylation. Level of glucose transporter 3 (GLUT3), a neuronal specific glucose transporter, is decreased in AD brain, which may contribute to impaired brain glucose uptake/metabolism. However, what causes the reduction of GLUT3 in AD brain is not fully understood. Here, we report 1) that decrease of GLUT3 is associated with the reduction of protein O-GlcNAcylation in AD brain, 2) that GLUT3 level is negatively correlated with calpain I activation in human brain, 3) that calpain I proteolyzes GLUT3 at the N-terminus in vitro, and 4) that activation of calpain I is negatively correlated with protein O-GlcNAcylation in AD brain. Furthermore, we found that overexpression of GLUT3 enhances protein O-GlcNAcylation in N2a cells. Overexpression of calpain I suppresses protein O-GlcNAcylation in these cells. These findings suggest a novel mechanism by which calpain I overactivation leads to GLUT3 degradation and the consequent down-regulation of protein O-GlcNAcylation in AD brain.

Keywords

Alzheimer’s disease calpain I glucose transporter 3 protein O-GlcNAcylation

INTRODUCTION

Alzheimer’s disease (AD) is characterized histopathologically by extracellular amyloid plaques and intracellular neurofibrillary tangles, which are mainly composed of abnormally hyperphosphorylated tau that aggregates into paired helical filaments [1]. Abnormal hyperphosphorylation of tau is critical to the molecular pathogenesis of AD and related tauopathies [2, 3].

In addition to phosphorylation, tau is also modified by O-GlcNAcylation, a post-translational modification by which the monosaccharide β-N-acetylglucosamine (GlcNAc) attaches to serine/threonine residues [4, 5]. In a previous study, we found that O-GlcNAcylation negatively regulates tau phosphorylation in a site-specific manner [5]. In AD brain, hyperphosphorylation of tau is negatively correlated with its O-GlcNAcylation [6]. O-GlcNAcylation of a protein is regulated by O-GlcNAc transferase, O-GlcNAcase, and the level of the donor substrate, UDP-GlcNAc [7]. Around 2–5% of the glucose is metabolized to UDP-GlcNAc through the hexosamine biosynthesis pathway [7]. Thus, the status of protein O-GlcNAcylation is influenced by intracellular level of glucose.

Glucose is the major source of energy for mammalian brain. Neurons cannot synthesize or store glucose. Glucose utilization/metabolism is impaired in AD brain [8 –11]. It is not well understood what causes the impairment of brain glucose metabolism in AD. Normally, the transport of glucose from the bloodstream into the brain is mediated by glucose transporters (GLUTs). Glucose transporter 3 (GLUT3) is a major neuronal GLUT that transports glucose into neurons [12 –14]. Reduced level of GLUT3 is correlated to tau hyperphosphorylation and density of neurofibrillary tangles in AD brains [15, 16], suggesting a possible role of GLUT3 in the impaired O-GlcNAcylation and tau pathology in AD.

Calpains are a family of cysteine proteases closely linked with AD [17]. They catalyze limited proteolytic cleavage of a variety of cellular proteins in all eukaryotes [18, 19]. Calpain I and calpain II are most predominantly and widely expressed in mammalian tissue, consisting of a large catalytic subunit (80 kDa) and a small regulatory subunit (28 kDa) [20]. They are presented principally as inactive precursors in the cell and are activated by Ca²⁺-stimulated autoproteolytic cleavage of the N-terminal sequence in response to Ca²⁺ influx. Calpain I and II are activated by micro-molar and milli-molar levels of calcium, respectively, and thus, they are also named μ-calpain and m-calpain. Calpain I, the major calpain isoform in the neuron, is truncated at the N-terminal and overactivated in AD brain [21, 22]. The role of over-activation of calpain I in the reduction of GLUT3 level in AD remains unknown.

In the present study, we show that GLUT3 overexpression enhances protein O-GlcNAcylation and the GLUT3 level correlates positively with protein O-GlcNAcylation and negatively with calpain I activation in human brain. Furthermore, we show that calpain I proteolyzes GLUT3 at the N-terminus and that overexpression of calpain I suppressed protein O-GlcNAcylation. These findings suggest a possible mechanism by which calpain I activation leads to reduction of GLUT3 and consequently the impairment of glucose uptake/metabolism and O-GlcNAcylation in AD brain.

MATERIALS AND METHODS

Plasmids, antibodies, and other reagents

pCI/GLUT3 tagged with haemagglutinin (HA) at N-terminus and Flag at C-terminus was generated by PCR amplification from pCMV/GLUT3 (Origene, USA) and its sequence was confirmed by DNA sequence analysis. Monoclonal and polyclonal anti-HA and polyclonal anti-calpain I were bought from Sigma (Sigma, St Louis, MO, USA). Pierce^TM O-linked N-Acetylglucosamine (RL2) antibody and ECL kit were from Thermo Scientific (Rockford, IL, USA). Anti-GLUT3 was from Chemicon (Temecula, CA, USA). Anti-GAPDH was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Horseradish peroxidase (HRP)-conjugated goat anti-mouse and goat anti-rabbit IgG were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA, USA). Oregon Green 488-conjugated goat anti-rabbit IgG, Alexa Fluor 555-conjugated goat anti-mouse IgG and Hoechst 33342 were from Invitrogen (Invitrogen, Carlsbad, CA, USA).

Human brain tissue

Medial frontal cortices of histopathologically-confirmed AD and age-matched normal human brains used in this study (Table 1) were obtained without identification of donors from the Sun Health Research Institute Donation Program (Sun City, AZ, USA). Brain samples were stored at –80°C until used. The use of frozen human brain tissue was in accordance with the National Institutes of Health guidelines and was exempted by the Institutional Review Board (IRB) of New York State Institute for Basic Research in Developmental Disabilities. The research does not involve intervention or interaction with the individuals. The information is not individually identifiable.

Table 1

Basic information on Alzheimer’s disease (AD) and control (Con) cases used in this study

Case	Age at death (y)	Gender	PMI^a (h)	Braak stage^b	Tangle scores^c
Con 1^d	85	M	2.5	II	4.25
Con 2^d	86	F	2.5	III	5.00
Con 3^d	81	M	2.75	III	6.41
Con 4^d	88	F	3.0	II	2.00
Con 5^d	90	F	3.0	III	4.50
Con 6^d	88	F	3.5	III	2.50
Con 7^d	88	F	3.0	IV	4.50
mean±SD	82.44±5.50		2.60±0.55		3.07±1.93
AD 1^d,e	89	F	3	V	14.5
AD 2^d,e	80	F	2.25	VI	14.5
AD 3^d,e	85	F	1.66	V	12.0
AD 4^d,e	78	F	1.83	VI	15.0
AD 5^d,e	95	F	3.16	VI	10.0
AD 6^d,e	86	M	2.25	VI	13.5
AD 7^d,e	91	F	3	V	8.50
Mean±SD	86.29±5.99		2.45±0.61		12.57±2.51
AD 8^e	83	F	3.0	VI	12.4
AD 9^e	74	M	2.75	VI	14.66
AD 10^e	79	F	1.5	VI	14.66
AD 11^e	73	F	2.0	V	15.00
AD 12^e	81	M	3.0	V	11.00
AD 13^e	76	M	2.33	VI	15.00
AD 14^e	72	M	2.5	VI	15.00
AD 15^e	74	F	2.83	VI	15.00
AD 16^e	76	M	4.0	V	15.00
AD 17^e	78	M	1.83	VI	15.00
AD 18^e	79	F	4.05	N/A	N/A
AD 19^e	86	F	3.92	N/A	N/A
AD 20^e	63	F	3.33	N/A	N/A
AD 21^e	97	F	3.25	N/A	N/A

^a PMI, postmortem interval. ^bNeurofibrillary pathology was staged according to Braak and Braak (1995). ^cTangle score was a density estimate and was designated none, sparse, moderate, or frequent (0, 1, 2, or 3 for statistics), as defined according to CERADAD criteria. Five areas (frontal, temporal, parietal, hippocampal, and entorhinal) were examined, and the scores were added up for a maximum of 15. N/A, not available. ^dCases were used for study in Fig. 2. ^eCases were used for study in Fig. 4.

Cell culture and transfection

HeLa cells and HEK-293FT cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (Invitrogen) at 37°C (5% CO₂). Mouse neuroblastoma N2a cells were maintained in DMEM/F12 medium with 10% FBS. Transfections were performed with Lipofectamine LTX (Invitrogen) in HeLa cells, with Fugene HD (Promega, Madison, WI, USA) in N2a and HEK-293FT cells according to the manufacturer’s instructions.

Immunofluorescence staining

HeLa cells were plated onto poly-D-lysine pre-coated coverslips in 24-well plate 1 day prior to transfection and transfected with pCI/HA·GLUT3·Flag by Lipofectamine LTX for 48 h. Then cells were washed with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde in PBS for 30 min at room temperature, blocked with 10% goat serum in 0.2% Triton X-100/PBS for 2 h at 37°C, and incubated with polyclonal anti-HA or together with RL2 antibody overnight at 4°C. After incubation with secondary antibody (Oregon Green 488-conjugated goat anti-rabbit IgG, Alexa Fluor 555-conjugated goat anti-mouse IgG), the cells were incubated with 5 μg/ml Hoechst 33342 for 15 min at room temperature and mounted with Fluoromount-G. The immunofluorescence was analyzed with a Leica TCS SP2 laser-scanning confocal microscope.

Western blots and immuno-dot blots

Human brain tissue was homogenized in 9×volumes of buffer containing 50 mM Tris-HCl, pH 7.4, 8.5% sucrose, 10 mM β-mercaptoethanol, 2.0 mM EDTA, 2.0 mM benzamidine, and 10 μg/ml each of aprotinin, leupeptin, and pepstatin. The brain homogenates were added 4×Laemmli sample buffer (125 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 10% β-mercaptoethanol, 0.004% bromphenol blue) followed by heating in boiling water for 5 min. Cultured cells were lysed in 1×Laemmli sample buffer and boiled in water for 5 min. The protein concentration was measured by Pierce^TM 660 nm protein assay kit (Thermo Scientific). Equal amounts of protein were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and were electrically blotted onto PVDF membrane. The blot was blocked with 5% fat-free milk in Tris-buffered saline (TBS, 50 mM Tris-HCl, pH 7.4, 150 mM NaCl) for 30 min and then incubated with primary antibody overnight at room temperature. After washing three times with TBST (TBS with 0.05% Tween 20), the blot was incubated with HRP-conjugated secondary antibody for 2 h. After being washed three times with TBST, the blot was incubated with ECL for 1 min and then exposed to X-ray film (Kodak, Rochester, NY, USA).

Immuno-dot blots were performed as described previously [6].

Immunoprecipitation of GLUT3

pCI/HA·GLUT3·Flag was transfected into HEK-293FT cells. The cells were washed twice with PBS, and lysed with RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% NP-40, 1 mM Na₃VO₄, 50 mM NaF, 2 mM EDTA, 1 mM PMSF, and 10 μg/mL of aprotinin, leupeptin, and pepstatin) on ice for 20 min, then centrifuged at 15,000×g, 4°C for 5 min. The supernatant was incubated with protein G beads (pre-coupled with monoclonal anti-HA) overnight at 4°C. The beads were washed with lysis buffer twice and with TBS twice. Immunoprecipitated GLUT3 was subjected for in vitro proteolysis.

In vitro proteolysis of GLUT3

Human brain homogenate was prepared as described above except the protease inhibitors were omitted from the homogenizing buffer and centrifuged at 15,000×g, 4°C for 10 min to obtain the crude extract. Then, the crude brain extract was incubated in the presence of EGTA (ethylene glycol-bis (β-aminoethyl ether)-N,N,N’,N’-tetraacetic acid) or various concentrations of Ca²⁺ and/or various selective protease inhibitors for 10 min at 30°C.

Immunoprecipitated GLUT3 as described above was incubated with different concentrations of calpain I (0-0.67 μg/ml) in reaction buffer (50 mM Tris-HCl, pH7.4, 0.5 mM CaCl₂) for 10 min at 30°C.

The proteolysis reaction was terminated by addition of 2×Laemmli sample buffer, followed by heating in boiling water for 5 min. The products of proteolysis were analyzed by western blots developed with indicated antibodies.

Statistical analysis

The data are presented as mean±SD. Data points were compared by the unpaired two-tailed Student’s t test (for data with normal distribution) or Mann Whitney test (for data with non-normal distribution). For analysis of the correlation between GLUT3 and O-GlcNAcylation or calpain I activation, Pearson correlation coefficient was calculated.

RESULTS

GLUT3 overexpression enhances protein O-GlcNAcylation

GLUT3 is the major neuronal glucose transporter. We therefore studied the effect of GLUT3 expression on protein O-GlcNAcylation in neuroblastoma N2a cells. To verify the cell membrane expression of GLUT3, we first transfected pCI/GLUT3 tagged with HA and Flag at the N- and C-termini, respectively, into HeLa cells, which have clear and large cytoplasm and nucleus, and immunostained with polyclonal anti-HA. We found that GLUT3 was mainly located on the cell membrane (Fig. 1A). To determine the role of GLUT3 in O-GlcNAcylation, we then transfected various amount of pCI/GLUT3 into N2a cells. The cells were maintained with DMEM/FBS containing different concentrations of glucose (Glc. 4.5 g/L or 1.0 g/L). We found that overexpression of GLUT3 increased O-GlcNAcylation in N2a cells dose-dependently under two glucose concentrations in the culture medium. The levels of O-GlcNAcylation were higher in cells cultured at high glucose concentration (4.5 g/L) (Fig. 1B, C). To further verify the relationship between GLUT3 and O-GlcNAcylation, we overexpressed GLUT3 tagged with HA and Flag in HeLa cells and immunostained the cells with polyclonal anti-HA and RL2 antibody. We found that the RL2 immuno-signal was higher in the cells with GLUT3 than that without GLUT3 overexpression (Fig. 1D), confirming that GLUT3 overexpression enhances protein O-GlcNAcylation.

Fig.1

GLUT3 enhances protein O-GlcNAcylation in cultured cells. (A) HeLa cells were transfected with pCI/HA·GLUT3·Flag for 48 h and immunostained with polyclonal anti-HA followed with Oregon Green 488-conjugated goat anti-rabbit IgG (green). Hoechst staining (blue) was used to stain nuclei. (B) N2a cells were transfected with various amounts of pCI/HA GLUT3 Flag for 48 h and maintained in cultured medium containing 1.0 g/L or 4.5 g/L glucose. The level of protein O-GlcNAcylation and the expression of GLUT3 were analyzed by western blots developed with RL2 (O-GlcNAc) and anti-HA (GLUT3), respectively. GAPDH blot was included as a loading control. (C) The relative levels of protein O-GlcNAcylation determined by RL2 were plotted against the corresponding amount of pCI/GLUT3 transfected into the cells. (D) GLUT3 tagged with HA was overexpressed in HeLa cells for 48 h. The cells were immnostained with polyclonal anti-HA and RL2 antibody followed with Oregon Green 488-conjugated goat anti-rabbit IgG (green) and Alexa Fluor 555-conjugated goat anti-mouse IgG (red). Hoechst staining (blue) was used to stain nuclei. Scale bar 25 μm.

Level of GLUT3 positively correlates with protein O-GlcNAcylation and negatively with Calpain I activation in human brain

To learn the possible role of GLUT3 in the downregulation of protein O-GlcNAcylation in AD, we determined the levels of GLUT3 in AD and control brains by western blots and of protein O-GlcNAcylation by immuno-dot-blots. We found reduction of GLUT3 in AD brains as compared to the age- and postmortem delay matched control brains (Fig. 2A, B). A reduction of O-GlcNAcylation in AD brain was found previously by using the same brain samples [5]. We performed Pearson Correlation analysis to learn if there is any relationship between GLUT3 and O-GlcNAcylation in the human brain. We found that the level of GLUT3 correlated positively with the level of O-GlcNAcylation (Fig. 2C). These results suggest that the decreased GLUT3 may be responsible for the impaired O-GlcNAcylation in AD brain.

Fig.2

Decreased GLUT3 is associated with low level of O-GlcNAcylation and overactivation of calpain I in AD brains. (A) Frontal cortices from 7 AD patients and 7 controls were analyzed by western blots for GLUT3, calpain I and GAPDH. ← indicates the full length calpain I and ] indicates truncated calpain I. (B) The level of GLUT3 determined by western blots in panel A was quantified by densitometry and normalized with GAPDH. (C) Level of GLUT3 (Y-axis) was plotted against the level of O-GlcNAcylation (X-axis) determined by immuno-dot blots. One control case circled was excluded for the correlation analysis. (D) Activation of calpain I (truncation/total) was calculated after densitometry. (E) Activation of calpain I was then plotted against the corresponding GLUT3 level. The Pearson correlation coefficient r was calculated. The data are presented as mean±S.D. **, p < 0.01. Blue symbol, control cases and red symbol, AD cases.

We found previously that calpain I (μ-calpain), one of major calpain isoform in the neuron, is truncated and activated in AD brain, and the overactivation causes truncation and degradation of several neuronal proteins [21, 22]. To learn whether calpain I overactivation also contributes to the reduction of GLUT3 in AD brain, we determined the activation of calpain I in the same brain samples by western blots and found that in AD brains calpain I was proteolyzed from the 80-kDa inactive form to 78-kDa and 76-kDa truncated and active forms (Fig. 2A, D). Correlation analysis revealed a very strong negative relationship between the GLUT3 level and the calpain I activation in human brains (Fig. 2E). These results suggest that calpain I activation might play a role in the reduction of GLUT3 in AD brain.

Calpain I proteolyzes GLUT3 at the N-terminus in vitro

To learn whether GLUT3 is proteolyzed by calpain I in human brain, we incubated normal human brain extract with various concentrations of Ca²⁺ (0–2.14 mM), and determined the truncation/activation of calpain I and the cleavage of GLUT3 by western blots. We found that accompanied by calpain I truncation from the 80-kDa to 78-kDa and 76-kDa, the level of GLUT3 was reduced in a Ca²⁺ dose-dependent manner (Fig. 3A), suggesting that GLUT3 is proteolyzed by a calcium-dependent protease(s).

Fig.3

GLUT3 is proteolyzed by Calpain I at the N-terminus in vitro. (A) Human brain extract was incubated with different concentrations of Ca²⁺ and analyzed by western blots developed with anti-calpain I and anti-GLUT3. (B) Human brain extract was incubated with EGTA, Ca²⁺ or together with various selective protease inhibitors and analyzed by western blots developed with anti-calpain I or anti-GLUT3. (C) GLUT3 tagged with HA at N-terminus and Flag at C-terminus was overexpressed in HEK-293FT cells and immunoprecipitated with monoclonal anti-HA. The immunopurified GLUT3 was incubated with different concentrations of calpain I and analyzed by western blots developed with antibodies against HA or Flag. ← indicates the full length calpain I and ] indicates truncated calpain I.

Calpain I is a calcium-dependent cystine protease and auto-proteolyzed and activated by micromolar concentration of calcium. We incubated the brain extract with EGTA, a calcium chelator (2 mM), Ca²⁺ (2 mM) alone or together with various selective protease inhibitors. We found that EGTA, chelated all Ca²⁺, suppressed the proteolysis of both calpain I and GLUT3, confirming the role of calcium in GLUT3 proteolysis. However, aprotinin, a serine protease inhibitor (10 μg/ml) and pepstatin, an aspartic protease inhibitor (10 μg/ml) could not inhibit the proteolysis of either calpain I or GLUT3 in the human brain extracts. In contrast, leupeptin, a selective inhibitor of cysteine and serine proteases (10 μg/ml), N-acetyl-Leu-Leu-Nle-CHO (ALLN), a calpain and cysteine protease inhibitor (0.5 or 5 μM), and calpastatin, the endogenous calpain specific inhibitor (0.1 or 1 μM) were found to inhibit proteolysis of calpain I and GLUT3 markedly (Fig. 3B). These results suggest that the degradation of GLUT3 in human brain extract was most likely due to calpain.

Fig. 3A shows that only micro-molar levels of free Ca²⁺ were needed for the truncation of GLUT3 present in the reaction mixture during incubation of the brain extract, suggesting that the GLUT3 proteolysis in the human brain extract resulted from activation of calpain I rather than calpain II, which requires milli-molar concentrations of free Ca²⁺ for activation. Taken together, these results suggest that GLUT3 is proteolyzed most likely by calpain I in the human brain.

To further investigate the proteolysis of GLUT3 by calpain I, we overexpressed GLUT3 with HA-tagged at the N-terminus and Flag-tagged at the C-terminus in HEK-293FT cells and immunopurified it with monoclonal anti-HA. The immunopurified HA·GLUT3·Flag was incubated with different concentrations of calpain I (0–0.67 μg/ml) and analyzed by western blots developed with antibodies against HA and Flag. We found that the levels of GLUT3 were decreased as detected by anti-HA and the GLUT3 was proteolyzed as detected by anti-Flag in a calpain I-concentration-dependent manner. These results suggest that calpain I proteolyzes GLUT3 at the N-terminus (Fig. 3C).

Activation of calpain I negatively correlates with O-GlcNAcylation in AD brains

The results presented above suggest a role of calpain I activation in the reduction of GLUT3 that could result in downregulation of O-GlcNAcylation. To learn the role of calpain I in protein O-GlcNAcylation, we overexpressed calpain I in N2a cells and analyzed protein O-GlcNAcylation by western blots. We found that protein O-GlcNAcylation was decreased significantly in N2a cells with calpain I overexpression (Fig. 4A, B).

Fig.4

Activation of calpain I is correlated with the protein O-GlcNAcylation negatively. (A) Calpain I was overexpressed in N2a cells for 48 h. The expression of calpain I and protein O-GlcNAcylation were analyzed by western blots. (B) The levels of protein O-GlcNAcylation were quantified by densitometry and are presented as mean±S.D. **, p < 0.01. (C) Frontal cortices from 21 AD cases were analyzed by western blots developed with anti-calpain I and RL2 (O-GlcNAc). ← indicates the full length calpain I and ] indicates truncated calpain I. (D) Activation of calpain I (truncated/total calpain I) was plotted against the corresponding O-GlcNAc level. Pearson correlation coefficient r was calculated.

We further analyzed by western blots the levels of calpain I activation and O-GlcNAcylation in the frontal cortices from 21 AD cases (Fig. 4C). We found that the level of activation of calpain I was negatively correlated with the level of protein O-GlcNAcylation in AD brains (Fig. 4D). These results strongly suggest the role of calpain I activation in the reduction of protein O-GlcNAcylation in AD brain.

DISCUSSION

Approximately 2–5% of total intracellular glucose feeds into the hexosamine biosynthesis pathway to produce glucosamine-6-phosphate and, ultimately, UDP-N-Acetylglucosamine (UDP-GlcNAc) [7]. As the major neuronal glucose transporter, GLUT3 controls neuronal glucose flux and regulates protein O-GlcNAcylation. In the present study, we found that overexpression of GLUT3 enhances protein O-GlcNAcylation, and that decreased GLUT3 is associated with reduction of protein O-GlcNAcylation and calpain I activation in AD brain. Calpain I proteolyzed GLUT3 in vitro at the N-terminus. Overexpression of calpain I suppressed protein O-GlcNAcylation. In AD brain, overactivation of calpain I correlated negatively with protein O-GlcNAcylation. Thus, it appears likely that overactivation of calpain I as a result of calcium overload in AD brain proteolyzes GLUT3, contributes to impairment of glucose uptake/metabolism, and leads to reduction of protein O-GlcNAcylation.

Impairment of glucose uptake/metabolism occurs prior to the appearance of clinical symptoms and mild cognitive impairment, a likely precursor of AD [23 –29], suggesting that impairment of brain glucose uptake/metabolism could be a cause, rather than a consequence, of AD. We previously reported that the impairment of glucose uptake/metabolism contributes to AD pathogenesis through downregulation of protein O-GlcNAcylation, one of post-translational modifications of protein [6]. O-GlcNAcylation modifies serine or threonine residues and regulates the phosphorylation of proteins. Tau is modified by O-GlcNAcylation, and O-GlcNAcylation regulates tau phosphorylation site-specifically [5]. In AD brain, O-GlcNAcylation is reduced and associated with tau hyperphosphorylation [5, 6]. In the present study, we found that O-GlcNAcylation is correlated with GLUT3 level in human brain. Overexpression of GLUT3 increased protein O-GlcNAcylation dose-dependently. Thus, decreased GLUT3 may contribute to tau pathogenesis through downregulation of O-GlcNAcylation in AD brain.

Glucose is the major energy substrate for mammalian brain metabolism. The transport of glucose from the bloodstream into the brain is mediated by GLUTs, and GLUT3 is the predominant GLUT responsible for glucose transport into neurons [12, 15]. We previously reported that GLUT3 is decreased in AD brains [15, 30]. The PKA-CREB signaling regulates GLUT3 expression. Calpain I proteolyzes CREB and regulatory subunits of PKA [30, 31]. Overactivation of calpain I downregulates PKA-CREB signaling, leading to reduction of GLUT3 expression [30]. In the present study, we found that calpain I also proteolyzes GLUT3 directly. Thus, overactivation of calpain I in AD brain may contribute to the reduction of GLUT3 directly through its degradation and indirectly through PKA-CREB signaling.

Calpain I is overactivated by truncation from 80 kDa full-length to 78-76 kDa active forms in AD brain [21, 22]. Overactivation of calpain I plays critical roles in AD pathogenesis [32]. Consistently, we confirmed an increase in the truncation of calpain I, indicating its overactivation in AD brain. Calpain I limited cleaves many proteins to truncated forms. Thus, we do not expect, but could not rule out, a significant impact of the cleavage on O-GlcNAc level, which remains elusive. In the present study, we found that accompanied with truncation of calpain induced by calcium, the level of GLUT3 in human extract was decreased dose-dependently on calcium, suggesting an association of GLUT3 with calcium/calpain I. Calpain I proteolyzed GLUT3 at the N-terminus. However, the exact cleavage sites of GLUT3 by calpain I remain to be identified. Overexpression of calpain I suppressed protein O-GlcNAcylation. Activation of calpain I is negatively correlated with the O-GlcNAcylation, suggesting that calpain I overactivation may contribute to tau pathology indirectly through O-GlcNAcylation. These findings provide a linkage between calpain I overactivation, GLUT3 reduction, and low level of O-GlcNAcylation in AD brain which are consistent with the hypothesis that impaired brain glucose uptake/metabolism contributes to neurodegeneration in AD. Inhibition of calpain I might be a potential therapeutic target for AD treatment.

Footnotes

ACKNOWLEDGMENTS

This work was supported in part by Nantong University and New York State Office for People with Developmental Disabilities and by grants from the National Natural Science Foundation of China (Grant 31671046), the U.S. Alzheimer’s Association (Grant DSAD-15-363172), Nantong Municipal Science and Technology project (MS12015058), China Scholarship Council (201608110204), and the Neural Regeneration Co-innovation Center of Jiangsu Province.

Authors’ disclosures available online ().

References

Grundke-Iqbal

, Iqbal

, Tung

, Quinlan

, Wisniewski

, Binder

(1986) Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A 83, 4913–4917.

Gong

, Liu

, Grundke-Iqbal

, Iqbal

(2005) Post-translational modifications of tau protein in Alzheimer’s disease. J Neural Transm (Vienna) 112, 813–838.

Iqbal

, Liu

, Gong

(2016) Tau and neurodegenerative disease: The story so far. Nat Rev Neurol 12, 15–27.

Arnold

, Johnson

, Cole

, Dong

, Lee

, Hart

(1996) The microtubule-associated protein tau is extensively modified with O-linked N-acetylglucosamine. J Biol Chem 271, 28741–28744.

Liu

, Iqbal

, Grundke-Iqbal

, Hart

, Gong

(2004) O-GlcNAcylation regulates phosphorylation of tau: A mechanism involved in Alzheimer’s disease. Proc Natl Acad Sci U S A 101, 10804–10809.

Liu

, Shi

, Tanimukai

, Gu

, Grundke-Iqbal

, Iqbal

, Gong

(2009) Reduced O-GlcNAcylation links lower brain glucose metabolism and tau pathology in Alzheimer’s disease. Brain 132, 1820–1832.

Love

, Hanover

(2005) The hexosamine signaling pathway: Deciphering the “O-GlcNAc code”. Sci STKE 2005, re13.

Heiss

, Szelies

, Kessler

, Herholz

(1991) Abnormalities of energy metabolism in Alzheimer’s disease studied with PET. Ann N Y Acad Sci 640, 65–71.

McGeer

, McGeer

, Akiyama

, Harrop

(1989) Cortical glutaminase, beta-glucuronidase and glucose utilization in Alzheimer’s disease. Can J Neurol Sci 16, 511–515.

10.

McGeer

, Peppard

, McGeer

, Tuokko

, Crockett

, Parks

, Akiyama

, Calne

, Beattie

, Harrop

(1990) 18Fluorodeoxyglucose positron emission tomography studies in presumed Alzheimer cases, including 13 serial scans. Can J Neurol Sci 17, 1–11.

11.

Smith

, de Leon

, George

, Kluger

, Volkow

, McRae

, Golomb

, Ferris

, Reisberg

, Ciaravino

, et al. (1992) Topography of cross-sectional and longitudinal glucose metabolic deficits in Alzheimer’s disease. Pathophysiologic implications. Arch Neurol 49, 1142–1150.

12.

McEwen

, Reagan

(2004) Glucose transporter expression in the central nervous system: Relationship to synaptic function. Eur J Pharmacol 490, 13–24.

13.

Benarroch

(2014) Brain glucose transporters: Implications for neurologic disease. Neurology 82, 1374–1379.

14.

Szablewski

(2017) Glucose transporters in brain: In health and in Alzheimer’s disease. J Alzheimers Dis 55, 1307–1320.

15.

Liu

, Liu

, Iqbal

, Grundke-Iqbal

, Gong

(2008) Decreased glucose transporters correlate to abnormal hyperphosphorylation of tau in Alzheimer disease. FEBS Lett 582, 359–364.

16.

Yan

, Zhang

, Wang

, Xu

, Ren

, Zhang

, Shi

, Chen

, Shi

, Tian

, Zhao

, Dong

(2014) Significant reduction of the GLUT3 level, but not GLUT1 level, was observed in the brain tissues of several scrapie experimental animals and scrapie-infected cell lines. Mol Neurobiol 49, 991–1004.

17.

Kurbatskaya

, Phillips

, Croft

, Dentoni

, Hughes

, Wade

, Al-Sarraj

, Troakes

, O’Neill

, Perez-Nievas

, Hanger

, Noble

(2016) Upregulation of calpain activity precedes tau phosphorylation and loss of synaptic proteins in Alzheimer’s disease brain. Acta Neuropathol Commun 4, 34.

18.

Mattson

, Chan

(2003) Neuronal and glial calcium signaling in Alzheimer’s disease. Cell Calcium 34, 385–397.

19.

Goll

, Thompson

, Li

, Wei

, Cong

(2003) The calpain system. Physiol Rev 83, 731–801.

20.

Pal

, Elce

, Jia

(2001) Dissociation and aggregation of calpain in the presence of calcium. J Biol Chem 276, 47233–47238.

21.

Liu

, Grundke-Iqbal

, Iqbal

, Oda

, Tomizawa

, Gong

(2005) Truncation and activation of calcineurin A by calpain I in Alzheimer disease brain. J Biol Chem 280, 37755–37762.

22.

Saito

, Elce

, Hamos

, Nixon

(1993) Widespread activation of calcium-activated neutral proteinase (calpain) in the brain in Alzheimer disease: A potential molecular basis for neuronal degeneration. Proc Natl Acad Sci U S A 90, 2628–2632.

23.

Pietrini

, Azari

, Grady

, Salerno

, Gonzales-Aviles

, Heston

, Pettigrew

, Horwitz

, Haxby

, Schapiro

(1993) Pattern of cerebral metabolic interactions in a subject with isolated amnesia at risk for Alzheimer’s disease: A longitudinal evaluation. Dementia 4, 94–101.

24.

Mielke

, Herholz

, Grond

, Kessler

, Heiss

(1994) Clinical deterioration in probable Alzheimer’s disease correlates with progressive metabolic impairment of association areas. Dementia 5, 36–41.

25.

de Leon

, Convit

, Wolf

, Tarshish

, DeSanti

, Rusinek

, Tsui

, Kandil

, Scherer

, Roche

, Imossi

, Thorn

, Bobinski

, Caraos

, Lesbre

, Schlyer

, Poirier

, Reisberg

, Fowler

(2001) Prediction of cognitive decline in normal elderly subjects with 2-[(18)F]fluoro-2-deoxy-D-glucose/poitron-emission tomography (FDG/PET). Proc Natl Acad Sci U S A 98, 10966–10971.

26.

Drzezga

, Lautenschlager

, Siebner

, Riemenschneider

, Willoch

, Minoshima

, Schwaiger

, Kurz

(2003) Cerebral metabolic changes accompanying conversion of mild cognitive impairment into Alzheimer’s disease: A PET follow-up study. Eur J Nucl Med Mol Imaging 30, 1104–1113.

27.

Drzezga

, Riemenschneider

, Strassner

, Grimmer

, Peller

, Knoll

, Wagenpfeil

, Minoshima

, Schwaiger

, Kurz

(2005) Cerebral glucose metabolism in patients with AD and different APOE genotypes. Neurology 64, 102–107.

28.

Mosconi

, Nacmias

, Sorbi

, De Cristofaro

, Fayazz

, Tedde

, Bracco

, Herholz

, Pupi

(2004) Brain metabolic decreases related to the dose of the ApoE e4 allele in Alzheimer’s disease. J Neurol Neurosurg Psychiatry 75, 370–376.

29.

Mosconi

, Tsui

, Rusinek

, De Santi

, Li

, Wang

, Pupi

, Fowler

, de Leon

(2007) Quantitation, regional vulnerability, and kinetic modeling of brain glucose metabolism in mild Alzheimer’s disease. Eur J Nucl Med Mol Imaging 34, 1467–1479.

30.

Jin

, Qian

, Yin

, Zhang

, Iqbal

, Grundke-Iqbal

, Gong

, Liu

(2013) CREB regulates the expression of neuronal glucose transporter 3: A possible mechanism related to impaired brain glucose uptake in Alzheimer’s disease. Nucleic Acids Res 41, 3240–3256.

31.

Liang

, Liu

, Grundke-Iqbal

, Iqbal

, Gong

(2007) Down-regulation of cAMP-dependent protein kinase by over-activated calpain in Alzheimer disease brain. J Neurochem 103, 2462–2470.

32.

Nixon

, Saito

, Grynspan

, Griffin

, Katayama

, Honda

, Mohan

, Shea

, Beermann

(1994) Calcium-activated neutral proteinase (calpain) system in aging and Alzheimer’s disease. Ann N Y Acad Sci 747, 77–91.