Comparison of β-Amyloid Plaque Labeling Methods:Antibody Staining, Gallyas Silver Staining,and Thioflavin-S Staining

Xinze Shi, Xuan Wei, Longze Sha, Qi Xu＊

1State Key Laboratory of Medical Molecular Biology and Neuroscience Center, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences & School of Basic Medicine Peking Union Medical College, Beijing 100005, China 2High School Attached to Beijing University of Technology,Beijing 100022, China

Key words: β-amyloid plaques; Alzheimer’s disease; antibody staining; Gallyas silver; thioflavin-S

ALZHEIMER’S disease (AD) is the most common neurodegenerative disorder and a major cause of dementia.1Its pathology shows β-amyloid (Aβ) deposition, tau hyperphosphorylation, brain atrophy, and neuronal death.2Evidence has been shown that Aβ is generated from neurotoxic amyloid oligomers and fibers. Aβ is derived from cleavage of amyloid precursor protein (APP) by beta secretase and gamma secretase, and spreads among the brain and biofluids such as blood and cerebrospinal fluid.3

Over the years, many pathological methods have been used for detection of Aβ or neurofibrillary tangles(NFT), such as Congo red, silver, immunohistochemical,and thioflavin-S staining.4-6Indeed, several comparative studies on different staining methods have been performed,7-8which focused on staining morphology of Aβplaques, time-cost issues, or experimental repeatability.To our knowledge, no study reported on sensitivity of the various staining methods. In the present study, we investigated spatiotemporal changes in Aβ plaque deposition in the brains of APP/presenilin 1 (PS1) mice at different ages. Accordingly, we compared the sensitivity of three different staining methods including antibody,thioflavin-S, and Gallyas silver staining to assess Aβ plaques. Further, we improved the experimental methods to minimize nonspecific reactions or fluorescence quenching, building on previous research.9

MATERIAlS AND METHODS

Animals

Different ages of B6C3-Tg (APPswe, PSEN1dE9) 85Dbo/Mmjax mice (34829-JAX, Jackson Laboratory, Bar Harbor, ME, USA), also known as APP/PS1 mice, were individually kept in cages with littermates and free access to water and food under 12-hour light/dark schedules.

All animal works were supported by the Experimental Animal Center of Peking Union Medical College,and in accordance with the institutional guidelines of the Beijing Administration Office of Laboratory Animals. All efforts were made to minimize animal suffering and reduce the number of animals used.

Anatomical processing

In total, 54 male APP/PS1 mice (aged from 2-12 months) were randomly divided into three groups:immunohistochemistry, modified Gallyas silver, and thioflavin-S staining groups respectively. After the mice were anesthetized, the brains were fixed by 4%paraformaldehyde perfusion through the heart. Paraffin-embedded blocks were prepared by sequential dehydration in graded ethanol before embedding and serial sectioning to a thickness of 4 μm. Sections were mounted on precoated glass slides.

Pathological staining

Dewaxed brain sections were rehydrated in distilled water and placed in a stainless-steel pressure cooker containing antigen retrieval solution. After 3 minutes of heating under pressure, sections were placed in a room-temperature water bath for 30 minutes, then washed twice in phosphate buffered saline (PBS) for 5 minutes each.

For immunohistochemical staining, 6E10 antibody(1:500; Biolegend, San Diego, CA, USA) was used because of its good specificity and sensitivity. The epitope lies within amino acids 3-8 of beta amyloid (EFRHDS).This antibody reacts to amino acids 1-17 of Aβ and detects abnormally processed isoforms. After incubation with 6E10 antibody at 4°C overnight, sections were treated with SP Rabbit & Mouse HRP Kit (CwBio,Beijing, China), according to the manufacturer’s protocol. Counterstaining was performed with hematoxylin(Golden Bridge Biotechnology Co., Ltd., Beijing, China)for the appropriate time. After each incubation step,sections were washed three times with tris-buffered saline with Tween 20 (TBST) for 5 minutes each.

For modified Gallyas Silver staining, deparaffinized slides were pre-incubated with 0.25% potassium permanganate solution for 5 minutes and 1% oxalic acid for 1 minute, followed by incubation in lanthanum nitrate-sodium acetate solution.10For induction of nucleation sites, sections were placed into an alkaline silver iodide solution for 30 minutes at 37°C. Next, sections were incubated in a physical developer solution for 15 minutes, rinsed in gold chloride solution, and then fixed in 1%thiosulfate solution. After each incubation step, sections were washed in distilled water for 5 minutes.

For thioflavin-S staining, modifications were made to minimize background staining based on numerous studies.11-12Like silver staining, sections were first pre-incubated with potassium permanganate solution and oxalic acid, then placed in 3% sodium borohydride solution for 5 minutes. To enhance the contrast between thioflavin-S and Aβ plaques, sections were incubated with 0.5% thioflavin-S dissolved in 50% ethanol for 30 minutes. This was followed by differentiation in two changes of 80% ethanol for 10 seconds each, and then incubation in 5 × PBS solution at 4°C for 1 hour.

Measurement of Aβ plaque and statistical analysis

After pathological staining, the number of Aβ plaques in the cortex and hippocampus was counted. Binary images were measured using ImageJ software (National Institutes of Health). As shown in Figs. 1-3, area occupied by Aβ plaques in the cortex and hippocampus was counted for each staining method for data analysis.To establish an accurate comparison of selectivity and sensitivity, adjacent brain sections were selected for Aβ plaques staining with each of the methods. As showed in Fig. 4, three areas of 1 mm×1 mm in each of three selected slices were analyzed. All photographs were captured with an inverted fluorescence microscope(DMI4000B; Leica, Wetzlar, Germany).

Prism software 7.0 (GraphPad Software Inc., La Jolla, CA, USA) was used to calculate statistical significance. One-way analysis of variance (ANOVA) followed by Tukey’s or Dunnett’s tests was used for multiple comparisons. Data are presented as mean±SEM. P values＜0.05 were considered statistically significant.

RESulTS

To examine Aβ plaque distribution at different ages of APP/PS1 mice, 6 different time points were chosen: 2, 4, 6, 8, 10, and 12 months (n=9 in each group), which represent most of a mouse’s lifespan.

First, 6E10 antibody immunohistochemistry assays were performed (Fig. 1A). Senile plaques first appeared in the cortex and hippocampus at 4 months,and then spread throughout the entire brain. Noticeably, Aβ deposition increased more rapidly with age in the cortex than the hippocampus. Quantification of area occupied by Aβ plaques in the cortex [Fig. 1B, F(5, 30)=41.35, P＜0.0001] and hippocampus [Fig. 1C,F (5, 30)=28.7, P＜0.0001] showed aggravated plaque formation with increasing age.

Figure 1. Distribution of β-amyloid plaque in the cortex and hippocampus of APP/PS1 mice. Senile plaque formation in 2, 4,6, 8, 10, and 12-month-old APP/PS1 mouse brain tissues detected by 6E10 immunohistochemistry staining (A). Scale bar in A: 200 mm. Quantification of area occupied by Aβ plaques in the cortex (B) and hippocampus (C) at different ages (n=3 for each group). Error bars represent mean±SEM. *P＜0.05, **P＜0.01, ***P＜0.001 compared with 2-month mice, based on one-way ANOVA followed by Dunnett’s test.

Gallyas silver staining is a classical argyrophil protein assay that is broadly used in neuropathy. We used a modified method that extends the incubation time of alkaline silver iodide solution, thereby increasing the staining effect. Senile plaques were detected at the earliest in 4-month-old brain (Fig. 2A), consistent with 6E10 immunohistochemistry results. As expected,fewer plaques were detected by Gallyas silver staining, particularly in aged mice. Likewise, senile plaques increased in the cortex [Fig. 2B, F (5, 30)=99.79,P＜0.0001] and the hippocampus [Fig. 2C, F (5,30)=26.5, P＜0.0001] with age.

Next, we examined Aβ plaques in the brain of APP/PS1 mice with thioflavin-S staining. A considerable number of plaques observed by 6E10 immunostaining and Gallyas silver staining were also detected by thioflavin-S staining (Fig. 3A). It is worth noting that many small background fluorescent dots were also detected,which could easily lead to a false positive result. Thus,to objectively evaluate thioflavin-S staining, we excluded this noisy signal using the in-built particle measure function of ImageJ. Similarly, thioflavin-S staining positive results improved in the cortex [Fig. 3B, F (5,30)=76.36, P＜0.0001] and hippocampus [Fig. 3C, F (5,30)=17.94, P ＜ 0.0001] with age.

Figure 2. Distribution of β-amyloid plaque in the cortex and hippocampus of APP/PS1 mice. Senile plaque formation in 2,4, 6, 8, 10, and 12-month-old APP/PS1 mouse brain tissues assayed by modified Gallyas silver staining (A). Scale bar in A: 200 mm. Quantification of area occupied by Aβ plaques in the cortex (B) and hippocampus (C) at different ages (n = 3 for each group). Error bars represent mean ± SEM. *P＜0.05, **P＜0.01, ***P＜0.001 compared with 2-month mice, based on one-way ANOVA followed by Dunnett’s test.

Figure 3. Distribution of β-amyloid plaque in the cortex and hippocampus of APP/PS1 mice. Senile plaque formation in 2, 4, 6, 8, 10, and 12-month-old APP/PS1 mouse brain tissue tested by modified thioflavin-S staining (A). Scale bar in A:200 mm. Quantification of area occupied by Aβ plaques in the cortex (B) and hippocampus (C) at different ages (n=3 for each group). Error bars represent mean ± SEM. *P＜0.05, ***P＜0.001 compared with two-month mice, based on one-way ANOVA followed by Dunnett’s test.

Subsequently, we compared staining results among 6E10 immunostaining, Gallyas silver staining,and thioflavin-S staining in contiguous sections of one mouse. For each slice, three 1 mm2areas were randomly selected, which corresponded to the other two staining methods (Fig. 4A-C). 6E10 immunostaining identified more positive plagues than Gallyas silver and thioflavin-S staining (P＜0.05). Moreover, there were no statistically significant differences between the latter two methods [Fig. 4D, F (2, 15)=8.559, P=0.0033].Antibody detection of Aβ was more sensitive than Gallyas silver and thioflavin-S staining.

DISCuSSION

Many pathological staining methods have been developed to describe pathological lesions related to neurological disorders. Although these methods have been used in research and clinical applications for decades, very few studies have compared the various methods. In this study, we investigated the spatiotemporal distribution of Aβ plaques in APP/PS1 transgenic mice and improved the staining quality by adjusting the procedure. We found that immunohistochemical staining is the most sensitive method for labeling Aβ plaques, compared with Gallyas silver staining and thioflavin-S staining.

Immunohistochemistry is an effective method to selectively detect proteins in tissues based on antibody antigen reaction, which is widely used in pathophysiology. Thioflavin-S or thioflavin-T can bind to the crossbeta-sheet structure of Aβ that is found in neuritic plaques, NFTs, neuropil threads, and vascular amyloid.13NFTs and Aβ plaques can be distinguished by fluorescence microscopy based on their shape. Studies have shown that thioflavin-T has higher permeability capacity through the blood-brain barrier than thioflavin-S.14Several attempts have been made to improve the sensitivity and reduce non-specific staining.15-16We extended the incubation time with a low concentration of thioflavin-S (0.5% in 50% ethanol) and high concentration of PBS to prevent fluorescence quenching.These improvements effectively reduced the influence of background staining (Fig. 3A). Several neuronal structures can be identified with different silver stains by using various solvent solutions or silver salts based on its argyrophil character.

Figure 4. Comparative analysis of β-amyloid plaques stained by different methods. Senile plaques in adjacent slices of the same APP/PS1 mouse stained by 6E10 immunohistochemistry (A), modified Gallyas silver staining (B), and modified thioflavin-S staining (C). Quantification of plaque density in the same region (D, n=9 for each group). Error bars represent mean±SEM. *P＜0.05, **P＜0.01 based on one-way ANOVA followed by Tukey’ s test. n.s.: not significant. Scale bar in A:200 mm.

Over the past two decades, studies have compared staining results of NFTs by various methods.In dog brain, previous studies have noted similar plague morphology between immunohistochemistry and various silver stains.17While another study in AD patient brain suggested that Campbell and Gallyas stain are superior to Bielschowsky or Congo red stain.18However, to date, there has been little discussion on comparisons within the same slice. Consecutive sections from the same mouse will obtain more reliable results (Fig. 4A-C). Further, no mature tangles around plaques have been observed in APP/PS1 mice, which avoids NFT disturbance.19Thus, in the present study, we have gone some way towards enhancing our understanding of amyloid plaque staining.

Conflicts of interest statement

The authors declare that they have no competing interests.