|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study was to look for alterations in β-amyloid peptide (Aβ) metabolism-related molecules in predementia, the early stage of Alzheimer’s disease (AD). AlCl3 (Al) and d-galactose (D-gal) were used to induce the mouse model for predementia and AD. Protein expression of β-amyloid (Aβ), β-secretase (BACE1), neprilysin (NEP), insulin degrading enzyme (IDE) and receptor for advanced glycation end products (RAGE) in the brain was measured. The results indicated that Al + D-gal induced an AD-like behavioral deficit at 90 days. The period from 45 to 75 days showed no significant behavioral deficit, and we tentatively define this as predementia in this model. A significant increase in BACE1 and decreasing NEP characterized days 45–90 in the cortex and hippocampus. However, high Aβ occurred at day 60. IDE increased from day 60 to day 75. There was no change in RAGE. The results suggest that the observed changes in BACE1, NEP and Aβ in predementia might relate to a different stage of the AD-like pathology, which may be developed into useful biomarkers for the diagnosis of very early AD.
Alzheimer’s disease (AD) is the most common cause of dementia in the elderly, accounting for 65–70% of all dementia (Blennow et al. 2006). However, early AD cannot be diagnosed even if the progressing cognitive deficits affect the patient’s ability to cope with the functional demands of his or her social and professional life (Hampel et al. 2008). Biomarkers are attracting increasing attention in research into early diagnosis of AD or as surrogate measures of the ongoing pathology. The major neuropathological hallmark of AD is senile plaque, which occurs with aging of the human brain many years prior to disease onset. Senile plaque is formed by conversion of the Aβ peptide to amyloid in the brain (Sommer 2002).
The overproduction or impaired clearance of cerebral β-amyloid peptide (Aβ) leads to elevated Aβ levels that promote Aβ aggregation, oligomerization, and fibrillogenesis (Hardy and Selkoe 2002). Aβ is generated by the cleavage of amyloid precursor protein (APP) by β-secretase (BACE1) and γ-secretase (Vardy et al. 2005). Pathways involved in the removal of cerebral amyloid peptide include enzymatic degradation and receptor-mediated clearance of the cerebral Aβ. The two major endopeptidases involved in Aβ enzymatic degradation are insulin degrading enzyme (IDE) and neprilysin (NEP). Multiple lines of evidence support a role for IDE and NEP in Aβ degradation (Farris et al. 2004; Tanzi et al. 2004; Wang et al. 2006). Because of the presence of a continuous monolayer of brain endothelial cells, the blood–brain barrier (BBB) in vivo does not allow free exchange of polar solutes such as Aβ between brain and blood. The receptor for advanced glycation end products (RAGE)—the specialized receptor for Aβ at the BBB—has been implicated strongly in amyloid peptide influx back into the central nervous system (CNS; Zlokovic 2004; Donahue et al. 2006). Therefore, the increased Aβ load in AD patients may result from the increased influx of circulating Aβ across the BBB and decreased Aβ degradation.
Most previous studies have focused on the examination of Aβ and Tau in cerebrospinal fluid (CSF) as markers in AD patients. Reports on other molecules of Aβ generation, transportation and degradation as markers for early diagnosis of AD are limited. Our previous study showed that Al + d-galactose (D-gal) induced an AD-like behavioral deficit at 90 days (Luo et al. 2008). In this study, the mouse was treated by Al + D-gal, and tested to see if molecules associated with Aβ metabolism were involved in the pathological process of AD before Aβ deposition in brain and the learning/memory injury of mice. We tentatively define the period from 45 to 75 days as predementia in this model, because there was no significant behavioral deficit during this period. Results show that alterations in the expression of BACE1, NEP and IDE occur earlier than high expression of Aβ in the mouse brain and behavior changes. This suggests that detection of BACE1, NEP and IDE levels as biomarkers could be useful for the early diagnosis of AD as well as representing potential pharmacological targets for the preservation of normal brain function and the treatment of AD.
Generation of the AD model (Luo et al. 2008): Kunming strain mice (6 to 8 weeks old and 20–25 g in groups of ten) were treated with AlCl3 40 mg kg−1 day−1 and D-gal 90 mg kg−1 day−1, i.p., once a day for 90 days using saline water as a vehicle. Animals were treated according to the guidelines of the Regulations of Experimental Animal Administration issued by the State Committee of Science and Technology of the People's Republic of China on 14 November 1988. The experiments were carried out under the approval of the Committee of Experimental Animal Administration of Nanjing University.
To evaluate the learning/memory ability of the mice, a water maze behavioral test was performed at 0, 45, 60, 75 and 90 days after the treatment, as described previously (Luques et al. 2007). A circular pool (150 cm in diameter and 45 cm deep) with walls and floor painted blank was filled with opaque water (23±1). A hidden circular platform (28 cm high, 12 cm in diameter, 2 cm below the water surface, fixed position) was located in the pool away from the pool wall. The pool was conceptually divided into four equal quadrants: NW, NE, SW and SE. Mice were trained on a visible platform (cured platform) for 3 days followed by 4 days of testing for learning and memory abilities with the platform submerged (hidden platform). Each mouse was subjected to four trials a day over a 4-day period. A mouse was placed in the water facing the pool wall at one of the four quadrants at a different place each day, and allowed to swim for a maximum of 90 s to locate the hidden platform, where it was allowed to stay for 10 s. The time taken for each mouse to locate the platform in all four trials was averaged and recorded. The daily administration of D-gal and Al was continued during the water maze performance tests.
The mice were given anesthesia at 0, 45, 60, 75 and 90 days after D-gal/Al treatment. Protein from the hippocampus and frontal cortex was extracted and then quantified using a Coomassie Blue Fast Staining Solution according to the manufacturer’s instructions. Equal amounts of protein samples were separated by SDS-PAGE and blotted onto polyvinylidene fluoride (PVDF) membranes (Xu et al. 2006). The membranes were probed with primary antibodies against Aβ1–42 (1:500, AB5078P, Chemicon, Temecula, CA), BACE1 (1:500, MAB5308, Chemicon), NEP (1:500, BAF1126, R&D), IDE (1:1,000, ab25970, abcam, Cambridge, UK), RAGE (1:500, AF1179, R&D Systems, Minneapolis, MN). GAPDH (1:500, Santa Cruz Europe, Heidelberg, Germany) was used as loading control. The proteins were detected using horseradish peroxidase-conjugated anti-rabbit or anti-mouse secondary antibodies and visualized using chemiluminescence reagents provided with the ECL kit (Amersham Pharmacia Biotech, Piscataway, NJ) and exposed to film. The intensity of blots was quantified by densitometry. All assays were performed at least in triplicate in independent experiments.
SPSS 13.0 software was used to handle all data. The results were expressed as mean ± SEM. Behavioral tests were determined by analysis of variance (one-way ANOVA) and paired t-test. The level of proteins was analyzed by Student’s t-test (two-tailed) or one-way ANOVA followed by post hoc Tukey test. The correlation between the expression of proteins and Aβ was assessed by Pearson correlation analyses. Statistical significance was set at P<0.05.
None of the mice tested had obvious health problems (e.g., weight loss, cataracts, or toxicity reactions). Throughout the training, on day 90, mice treated with Al and D-gal took longer to reach the platform than those of the vehicle group (P>0.05, Fig. 1). There was no significant difference between treatment and vehicle at 0, 45, 60, and 75 days. Each group consisted of ten mice.
To investigate if molecules of the associated Aβ pathway start to be induced before the decline of cognition function in this model, Aβ expression was measured by western blotting at 0, 45, 60, 75 and 90 days. The results indicated that there was a tendency for increased Aβ expression in the cortex (Co) and hippocampus (Hi) of the treated mice brain at 45 days, but this was not significant compared to the vehicle group. In contrast, a significant increase in Aβ expression was observed at matched time points on days 60, 75 and 90 (Figs. 2, ,3).3). Next, the level of BACE1 expression in the brain of the “predementia mouse” was examined. Compared to Aβ expression, high BACE1 appeared early. The level of BACE1 in the treatment group increased significantly from day 45 to 90 in both Co and Hi compared to control levels (Figs. 2, ,4).4). Statistical analysis showed that the level of BACE1 correlated with the level of Aβ (R=0.9351, P<0.05 for Hi; R=0.9688, P<0.01 for Co). The highest level was seen on day 75, confirming that overproduction of APP protein was not induced by Al and D-gal (data not shown). In this paper, no data for vehicle only is given because there was no difference between the treatment group at day 0 and vehicle groups from day 0 to day 90, thus we used day 0 as a control.
NEP and IDE are the main degrading enzymes for Aβ clearance in the brain. The protein levels of NEP and IDE were measured in the hippocampus and the frontal cortex. The results indicated that the level of NEP was significantly reduced in both tissues at day 45 (P<0.05) compared to that of the vehicle group, and decreased further at 60, 75 and 90 days (Figs. 2, ,5).5). However, the level of NEP did not correlate significantly with the level of Aβ (r=−0.8718, P>0.05 for Hi, r=−0.793, P>0.05 for Co). However, as Fig. 6 shows, expression of IDE in the cortex was significantly higher in the treatment group than that in the vehicle group at days 60 and 75 (P<0.05), and high expression of IDE in the hippocampus was observed only at day 60 (P<0.05). There was no significant change in IDE protein between the two groups at other time points (P>0.05).
RAGE mediates a continuous influx of circulating Aβ into the brain. To confirm whether RAGE is a useful marker for early diagnosis of AD, RAGE protein levels were determined. We found that expression of RAGE in the brain was a slightly higher in comparison with the control after treatment with Al + D-gal, but this difference did not reach significance (P<0.05, Figs. 2, ,77).
Alzheimer’s disease is currently most commonly identified using clinical criteria and psychometric cognitive assessment. The Mini-Mental State Examination (MMSE) is often used as a screening test for dementia. However, the MMSE has been found to be insensitive to early AD, and new tools for diagnosing early stages of AD are being developed (Benson et al. 2005; Nasreddine et al. 2005). Furthermore, in the early stage, there is no clinical method to determine whether early stage AD patients will progress to AD—long clinical follow up is required. Recently, some longitudinal studies have used PET or MRI images to predict and monitor cognitive decline from normal aging to early AD (Fritzsche et al. 2008; Mosconi et al. 2008). Both CSF biomarker levels and atrophy as assessed by MRI are now used in the diagnostic work-up of AD (Fox et al. 2005; Wiltfang et al. 2005; Waldemar et al. 2007). Increasing numbers of reports show that the amount of Aβ in the brain is determined by the rate of Aβ generation versus clearance. On this basis, it was hypothesized that alteration in the expression of Aβ-metabolism-associated molecules before pathological and behavioral changes could be developed into early markers of AD progression. The main findings of the present study were: (1) accumulation of high levels of BACE1 and Aβ in mouse brain occurred earlier than mouse memory impairment, (2) an increase in BACE1 and decrease in NEP were detected earlier than changes in Aβ, (3) the level of RAGE remained constant throughout the entire experimental period. This finding suggests that, besides Aβ, Aβ-pathway-related molecules such as BACE1 and NEP could also be developed into critical biological markers for early diagnosis of AD in the future.
The period from day 45 to day 75 after treatment with Al and D-gal was arbitrarily defined as the “predementia stage” in our current model. AD is seen to appear by day 90, because the behavioral test showed that the learning/memory of mice in the treatment group decreased significantly compared to that in the vehicle group at this time. The present study found that expression of BACE1 increased from day 45, i.e., earlier than Aβ expression, which begins to increase at day 60. Both BACE1 and Aβ alteration occurred before behavior changes. BACE1 cleavage is a pre-requisite for Aβ formation, and is the initiating step in Aβ generation. Ablation of BACE1 in transgenic AD models abolishes Aβ production and prevents the subsequent development of amyloid-associated pathologies (Ohno et al. 2004, 2007; McConlogue et al. 2007). BACE1 levels are elevated in both AD experimental models and in the AD patient brain (Li et al. 2004; Harada et al. 2006; Zhao et al. 2007; Tamagno et al. 2002, 2005). BACE1 activity is correlated with brain Aβ production in the frontal cortex (Li and Sudhof 2004; Gustaw et al. 2008), In agreement with others, our data suggests that BACE1 levels may start building early and become sustained during the course of AD development. In addition, the BACE1 elevation in AD may be actively involved in AD pathology and may occur prior to the appearance of overt neuron death (Zhao et al. 2007). Statistical analysis indicates that changes in BACE1 have a positive correlation with changes of brain Aβ.
Although the cause of increased Aβ in Alzheimer's disease can be attributed to many factors, evidence is accumulating that it may be a result of a decrease in Aβ clearance (Selkoe 2001). Aβ clearance occurs primarily through the action of NEP and IDE. The deletion or disruption of either of these peptidase genes in mice leads to a significant increase in Aβ levels (Farris et al. 2003; Miller et al. 2003). This study observed that NEP began to decrease at day 45. NEP changes occurred 15 days prior to the Aβ increase in these mouse brains. NEP cleaves and removes Aβ and, consequently, leads to a delay in amyloid plaque deposition in the Aβ-mediated AD brain (Mohajeri et al. 2002). NEP levels were reduced in affected AD brains, and in the brains of aged mice, and are known to be down-regulated in areas vulnerable to Aβ peptide accumulation in the AD brain (Yasojima et al. 2001; Iwata et al. 2002). This study and others suggest that NEP has a strong negative correlation with Aβ aggregation and thus may be a suitable candidate for development into a critical marker for the diagnosis of early AD.
On the contrary, levels of IDE, the rate-limiting Aβ-degrading enzyme (Hama et al. 2001), were significantly higher in the treatment group than in the vehicle group at day 60 in both Co and Hi, and at day 75 only in Co. The capacity of IDE to degrade Aβ has been recognized for many years (Kurochkin and Goto 1994). The cause of increased IDE may be due partly to activated astrocytes as a result of Aβ-triggered neuroinflammation (Leal et al. 2006). Contradictory results regarding IDE expression in the AD brain have been reported (Kim et al. 2007). Therefore IDE is not suitable as a special marker for the early diagnosis of AD since it might be involved in the inflammation reaction in the early stage of AD, and its role on Aβ degradation occurs after AD. The decrease in IDE happens in the late stage of AD (Tanzi et al. 2004).
RAGE has been characterized as a transporter of Aβ from the plasma to the brain across the BBB (Deane et al. 2004). RAGE is regarded as a multiligand receptor in the immunoglobulin (IgG) superfamily. RAGE binds soluble Aβ and mediates pathophysiologically relevant cellular responses (Stern et al. 2002). This study found no changes of RAGE levels in the brain at any timepoint in this model. In the central nervous system, RAGE is present at high levels during development, and its levels decline during maturity in normal aging (Leclerc et al. 2007). Microvascular RAGE expression was notably up-regulated in advanced AD as compared to early AD and control samples (Miller et al. 2008).
In conclusion, Al + D-gal treatment can generate high Aβ induction in mice brains, which serves as an animal model for predementia or AD. BACE1 and NEP alterations were apparent before changes in Aβ, showing positive and negative correlations with Aβ, respectively. Both may be developed into useful biomarkers for the early diagnosis of AD, and can also be regarded as targets for a therapeutic strategy against predementia and AD.
This work was supported by the National Nature Science Foundation of China (30470612, 30670739), the Doctoral Program Foundation of the Ministry of Education of China (20060284044), the International Cooperation Program and talented man program (BZ2006045, 06-B-002, RC2007006) of Jiangsu Province of China of Jiangsu Province of China, and 973 funding from the Ministry of Science and Technology in China (2009CB21906). We thank Marilyn White and Prof. Ying-Dong Zhang for writing modification.
Zong-Zheng Sun and Zhi-Bin Chen contributed equally to this work.