SEARCH

SEARCH BY CITATION

Keywords:

  • Alzheimer's disease;
  • amyloid β;
  • mass spectrometry;
  • secretase;
  • transgenic mice

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

Alzheimer's disease (AD) is marked by the presence of neurofibrillary tangles and amyloid plaques in the brain of patients. To study plaque formation, we report on further quantitative and qualitative analysis of human and mouse amyloid β peptides (Aβ) from brain extracts of transgenic mice overexpressing the London mutant of human amyloid precursor protein (APP). Using enzyme-linked immunosorbant assays (ELISAs) specific for either human or rodent Aβ, we found that the peptides from both species aggregated to form plaques. The ratios of deposited Aβ1–42/1–40 were in the order of 2–3 for human and 8–9 for mouse peptides, indicating preferential deposition of Aβ42. We also determined the identity and relative levels of other Aβ variants present in protein extracts from soluble and insoluble brain fractions. This was done by combined immunoprecipitation and mass spectrometry (IP/MS). The most prominent peptides truncated either at the carboxyl- or the amino-terminus were Aβ1–38 and Aβ11–42, respectively, and the latter was strongly enriched in the extracts of deposited peptides. Taken together, our data indicate that plaques of APP-London transgenic mice consist of aggregates of multiple human and mouse Aβ variants, and the human variants that we identified were previously detected in brain extracts of AD patients.

Abbreviations used
AD

Alzheimer's disease

AEC

3-amino-g-ethylcarbazole

APP

amyloid precursor protein

APPld

London mutant APP

APP-CTFβ1

APP-carboxyl-terminal fragment β1

BSA

bovine serum albumin

CHCA

α-cyano 4-hydroxy cinnaminic acid

DMSO

dimethyl sulphoxide

FCS

fetal calf serum

GuHCl

guanidine hydrochloride

amyloid β peptide

IP/MS

immunoprecipitation and mass spectrometry

MALDI-TOF

matrix-assisted laser desorption ionization time of flight

PBST

phosphate-buffered saline + 0.1% Triton X-100

pyroGlu

pyroglutamate

Alzheimer's disease (AD) is marked by the accumulation of neurofibrillary tangles and amyloid plaques in the brain. Plaques consist of aggregated amyloid β (Aβ) peptides that are produced by proteolytic cleavages of the amyloid precursor protein (APP) by β- and γ-secretase. As a result of APP processing, neurones produce Aβ peptides of different sizes, the most prominent being 40 or 42 amino acids long. The role of the secreted Aβ peptides in AD is still unclear, but many different genetic point mutations that lead to familial AD result in early Aβ aggregation. In most cases, the mutations directly affect APP processing towards Aβ, strongly implicating Aβ peptides with the disease. The Swedish double mutation in APP (KM670/671NL) mediates increased production of Aβ by enhanced β-secretase cleavage of the mutated precursor protein. The London mutation occurs at a different APP site (V717I), and shifts the balance of γ-secretase cleavage toward the 42 residue Aβ, which is more prone to aggregation than its shorter counterpart (reviewed by Selkoe 1999).

To study the role of Aβ in AD, multiple transgenic mouse strains have been developed that overexpress mutant human APP. In several mouse models of AD, the animals contain amyloid plaques in their brain, but this amyloidosis is not accompanied by extensive tangle formation or massive neuronal loss as in AD patients (reviewed by Janus et al. 2000). Nonetheless, the transgenic mice are considered a valid model to study amyloidosis and its pharmacological intervention. The aim of this study was to further analyse mice overexpressing London mutant human APP. The mice used in this study have been characterized previously. It was demonstrated immunohistochemically (IHC) that by 12–14 months of age the animals start developing plaques, which consist predominantly of deposited Aβ42 (Moechars et al. 1999). By 2 years of age, there is evidence for the existence of diffuse plaques, neuritic plaques and vascular amyloid aggregation. Contrary to the plaque Aβ, the vascular deposits consisted of an excess of Aβ40 (Van Dorpe et al. 2000).

In this report, we extended these studies by performing a detailed characterization of the Aβ peptides that are produced and deposited in the mice. We confirmed that the most prominent Aβ variants in plaques of APPld mice were Aβ1–42 and to a lesser extent Aβ1–40. More importantly, we demonstrated that the deposits contained significant levels of Aβ1–38 and Aβ11–42, and we showed that endogenous mouse Aβ coaggregated with human Aβ to form plaques.

Preparation of mouse brain extracts

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

All transgenic mice used in this study were of the FVB/N genetic background and expressed APP/V717I under the control of the thy1 gene promoter (Moechars et al. 1999). Experimental protocols were approved by the Institutional Review Committee of Janssen Pharmaceutica (Beerse, Belgium), and meet the European and Belgian guidelines on animal experimentation. Heterozygous APPld mice were anaesthetized with an overdose of Nembutal (15 mg per animal, injected intraperitoneally), and brains were isolated after perfusion with 15 mL cold saline. Brains were snap-frozen on liquid nitrogen and stored at −70°C.

Brains were thawed submerged in ice-cold buffer A [50 mm Tris pH 8.0, 150 mm NaCl, Complete protease inhibitor cocktail (Roche, Mannheim, Germany), and 0.8 µg/mL pepstatin (Roche)] and homogenized in a final volume of 2.5 mL buffer A, by 10 strokes at 1000 rpm in a potter homogenizer. The homogenates of brains from two littermates of the same gender were mixed and centrifuged (135 000 g in a MLA-80 rotor, 1 h at 4°C). The supernatant was collected and used for analysis of soluble Aβ by IP/MS. The pellet was resuspended in 1 mL guanidine hydrochloride (GuHCl) solubilization buffer (50 mm Tris pH 8.0, 6 m GuHCl) and sonicated for 10 s with a sonicator needle (Biobloc Scientific, Cergy, France). After 15 min incubation on ice, sonication was repeated for 5 s and the homogenate was diluted sixfold with ice-cold buffer A to a final GuHCl concentration of 1 m. Subsequent to a second centrifugation (135 000 g in a MLA-80 rotor, 1 h at 4°C), the supernatant was collected and used for analysis of deposited Aβ by enzyme-linked immunosorbent assay (ELISA) and immunoprecipitation and mass spectrometry (IP/MS).

ELISA quantification of soluble and deposited Aβ

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

Quantification of brain Aβ levels was done by sandwich ELISA. Human and rodent Aβ standards (American Peptide Company, Sunnyvale, CA, USA) used in ELISA and IP/MS were dissolved in dimethyl sulphoxide (DMSO) at 1 mg/mL, and further diluted to 5 µg/mL in buffer EC [20 mm sodium phosphate, 2 mm EDTA, 400 mm NaCl, 0.2% bovine serum albumin (BSA), 0.05% CHAPS, 0.4% casein, 0.05% NaN3, pH 7.0] and stored at −70°C. For use in ELISA, standards were diluted in casein buffer [0.1% casein in phosphate-buffered saline (PBS)] down to 2 pg/mL. 96-well plates were coated with monoclonal antibodies (100 µL at 5 µg/mL) that specifically recognize human and rodent Aβ ending either at residue 40 (JRF/cAβ40/10) or 42 (JRF/cAβ42/26) (Mercken et al. 2000; Vandermeeren et al. 2001). Prior to incubation in coated ELISA plates, GuHCl brain extracts were pre-diluted in casein buffer 10-fold or 300-fold, depending on subsequent use in rodent or human specific ELISAs, respectively. After overnight incubation at 4°C, horseradish peroxidase (HRP)-coupled antibodies were applied to washed plates. For selective detection of either rodent or human Aβ we used our respective species-specific monoclonal antibodies, JRF/rAβ/2 and JRF/AβN/25 (Mercken et al. 2000; Vandermeeren et al. 2001). Assays were developed with TMB/H2O2 substrate (Pierce, Rockford, IL, USA) according to the manufacturer's specifications.

IP/MS analysis

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

For immunoprecipitation of Aβ from brain homogenates, we used 500 µL NaCl extract or 250 µL GuHCl extract supplemented to 1 mL with Dulbecco's modified Eagle's medium + 1% fetal calf serum (FCS), added 3 µg antibody and protease inhibitors from a 10× concentrated solution of Complete cocktail. Antibodies used in this study were JRF/rAβ/2, JRF/AβN/25, JRF/cAβ40/10, JRF/cAβ42/26, or 4G8 (Senetek, Maryland Heights, MO, USA). After overnight incubation on a rotating platform at 4°C, 6 µL Protein G-speharose (50% slurry) was added and the incubation was continued for 3 h. Subsequently, beads were pelleted for 3 min at 1700 g in a cooled eppendorf centrifuge and washed sequentially with 1 mL PBS and 2 × 1 mL 5 mm HEPES pH 7.5. After the final wash, 5 µL buffer was left on top of the beads. For mass analysis of immunoprecipitated Aβ, 1.5 µL antibody-sepharose slurry was applied directly onto a matrix-assisted laser desorption ionization (MALDI)-target plate, the droplet was mixed on the spot with 1 µL saturated CHCA (α-cyano 4-hydroxy cinnaminic acid) matrix dissolved in 50% acetonitrile, 0.1% trifluoroacetic acid, and allowed to dry. Mass spectra were acquired on a Voyager DE Pro (Applied Biosystems, Foster City, CA, USA) in the positive ion linear mode. External calibration was done with respect to synthetic Aβ standards (American Peptide Company). Mass spectra were usually recorded up to 6000 Da and for some samples with high Aβ levels, up to 10 000 Da. As no specific peak was observed beyond 5000 Da, IP/MS traces shown in Fig. 2 were electronically cut off to increase peak resolution.

image

Figure 2. MALDI-TOF mass spectroscopy of Aβ peptides from brains of APPld mice. Aβ peptides were extracted from brains of APPld transgenic mice by sequential homogenization of brain tissue in NaCl buffer and GuHCl buffer as described in Materials and methods. Epitopes for antibodies used in IP/MS (a). Amino acids that differ between human and mouse Aβ are shown in bold. IP/MS from NaCl extracts using antibodies 4G8 (b) and JRF/AβN/25 (c). IP/MS from GuHCl extracts using antibodies 4G8 (d), JRF/Abβ/25 (e) and JRF/rAβ/2 (f). Peaks marked with an asterisk (*) were double charged background signals also observed in our control IP/MS with anti-KLH monoclonal antibodies. Peaks marked with a question mark were observed only upon IP/MS with JRF/AβN/25, but the mass did not correspond with that of any Aβ variant starting at residue Asp1.

Download figure to PowerPoint

Immunohistochemical analysis of APPld brain slices

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

Brain hemispheres from 5- and 16-month-old APPld mice were fixed overnight by immersion fixation with paraformaldehyde (4%) and cut into 50 µm vibratome sections, which were stored in PBS with 0.1% NaN3. Sections were washed two times in PBS before endogenous peroxidase blocking in 3% H2O2 for 30 min. After 3 × 5 min in PBS + 0.1% Triton X-100 (PBST) we incubated 20 min with 3% normal serum. The primary antibody was applied overnight in PBST with 1% normal serum. We then successively applied secondary antibody in PBST (60 min), ABC (Vector Laboratories, Burlingame, CA, USA) 1/100 in PBST (30 min) and AEC substrate (Vector; ± 20 min), with PBST (PBS before AEC) wash steps between each incubation. The sections were then mounted on gelatine-coated slides and dried for at least 45 min. Counterstaining was performed with Mayer's haematoxylin (Sigma, St Louis, MO, USA) and slides were mounted with Kaiser's Glycerol Gelatin (Fluka, Bornem, Belgium). Images were recorded at 10-fold magnification on a Leica microscope.

Deposition of human and mouse Aβ in APPld mice

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

To determine the levels of human and murine Aβ deposited in brains of APPld mice we used ELISAs that were developed in house. The insoluble Aβ peptides were extracted from brain homogenates using 6 m GuHCl. In homogenates from 12-month-old mice, human Aβ levels were relatively low (data not shown), confirming earlier conclusions that Aβ accumulation in APPld transgenic mice starts at 12–14 months of age (Moechars et al. 1999). Compared to the 1-year-old mice, Aβ levels increased by at least two orders of magnitude in extracts from 24-month-old mice. At this age, we found that Aβ concentrations varied significantly for different animals, especially when comparing mice of different gender. In males, Aβ deposition appeared to be considerably less than in females, the latter containing on average three times more Aβ1–40 and 1–42 (Table 1). Contrary to the absolute levels, ratios of deposited human Aβ42/Aβ40 were similar for both genders, ranging from 2.1 to 2.7 (Table 1).

Table 1.  Aβ levels in GuHCl brain extracts from 24-month-old APPld mice (µg peptide/g brain)
 Human Aβ1–40Human Aβ1–42Mean of ratios 42/40Mouse Aβ1–40Mouse Aβ1–42 Mean of ratios 42/40
RangeMean ± SDRangeMean ± SDRangeMean ± SDRangeMean ± SD
Eight females3.8–5.64.4 ± 0.87.4–13.49.5 ± 2.82.1 ± 0.30.06–0.100.08 ± 0.020.29–0.920.65 ± 0.328.3 ± 2.9
Six males1.0–1.71.3 ± 0.43.0–3.43.3 ± 0.22.7 ± 0.50.02–0.030.02 ± 0.010.12–0.260.13 ± 0.109.1 ± 2.6

To investigate whether endogenous murine Aβ was co-deposited with the transgene-derived human Aβ, we performed ELISAs specific for rodent Aβ. Prior to significant plaque formation (12 months), total murine Aβ levels were in the order of 1–3 ng/g transgenic brains, being similar to those of control mice (data not shown). In older APPld mice, the brain content of insoluble mouse Aβ increased dramatically, reaching up to 0.9 µg/g at 24 months of age (Table 1). The highest amounts were consistently detected for those mice that contained the most deposited human Aβ. This was reflected also by the fact that extracts from female animals had a higher load of mouse Aβ than their male counterparts (Table 1). Strikingly, the extracts contained up to 10 times more mouse Aβ1–42 than mouse Aβ1–40, suggesting that mouse Aβ42 was preferentially accumulating in plaques.

To show that the mouse Aβ in GuHCl extracts was in fact solubilized from plaques, we performed immunohistochemical staining on brain slices of transgenic mice. IHC with our rodent-specific antibody JRF/rAβ2 revealed the characteristic plaque staining in brain cortex from 16-month-old mice (Fig. 1a). In the same tissue, plaques were also observed using our human-specific antibody JRF/AβN/25 (Fig. 1b). On the other hand, IHC on brain slices obtained prior to plaque formation was negative with both antibodies (data not shown).

image

Figure 1. IHC analysis of deposits in brains of APPld mice. The presence of human and mouse Aβ in brain deposits in APPld transgenic mice was demonstrated by IHC on cortical sections, as described in Materials and methods. Selective staining of mouse and human Aβ was done using JRF/rAβ/2 (a) and JRF/AβN/25 (b) as primary antibodies, respectively.

Download figure to PowerPoint

Brains of APPld mice contain multiple Aβ variants

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

To assess whether Aβ peptides other than 1–40 and 1–42 were produced and deposited in the APPld mice, we performed IP/MS analysis on brain extracts of soluble and insoluble peptides. Secreted soluble Aβ was obtained by first homogenizing the brains in a buffer containing neither detergents nor chaotropic agents (referred to as NaCl extracts). Sequential to that, the insoluble Aβ peptides were extracted using 6 m GuHCl (GuHCl extracts). We consider the MS data as semiquantitative, and used the relative peak intensities within one spectrum to give an indication of the Aβ composition that was immunoprecipitated. However, ionisation efficiencies can vary significantly for different peptides. For example, in our set-up, an equimolar mixture of synthetic human Aβ1–40 and Aβ1–42 yielded a 30–40% higher signal for the shorter peptide. Immunoprecipitation of Aβ from brain extracts was done using antibodies that recognize different epitopes (Fig. 2a), some of which were also used in our ELISAs.

The sensitivity of our IP/MS analysis was insufficient to detect peaks in extracts from control mice. Representative IP/MS results from homogenates of APPld transgenic mouse brains are summarized in Fig. 2 and Table 2. Figures 2(b and c) show the analysis of soluble peptides from 24-month-old transgenic mice after IP with 4G8 or JRF/AβN/25, respectively. In addition to Aβ1–40, the peptides Aβ1–42 and Aβ1–38 were prominently present and weaker signals for Aβ1–37 and Aβ1–39 were also observed. Peak proportions were similar when using either 4G8 or JRF/AβN/25 for immunoprecipitation (Table 2), underlining the reproducibility of the results. On the other hand, only JRF/AβN/25 precipitated peptides ranging from Aβ1–13 to Aβ1–20, because these truncated Aβ variants lack the epitope for 4G8.

Table 2.  Peak data of spectra from IP/MS on brain extracts of APPld transgenic mice
IP/MS antibodyMr obsMr calcAβ variantRelative height (%)
  1. Observed (obs) and calculated (calc) masses correspond to protonated peptides of human (h) and murine (m) origin.

4G8, NaCl extract4075.34075.5h1–3717
4132.54132.6h1–3889
4232.14231.7h1–3925
4330.94330.9h1–40100
4515.24515.1h1–4274
 1562.11562.6h1–1344
1699.21699.7h1–1428
1717.9 ?80
1854.0 ?30
1955.21956.0h1–1620
 2068.22069.1h1–1731
JRF/AβN/25, NaCl extract2165.02168.3h1–1818
2314.62315.5h1–1929
2461.52462.7h1–2022
4073.14075.5h1–3715
4132.54132.6h1–3871
4231.64231.7h1–3921
4330.94330.9h1–40100
4515.04515.1h1–4269
 3318.33318.9pyroGlu h11–427
3336.23336.9h11–4219
3355.33356.0m11–4210
4G8, GuHCl extract4074.44075.5h1–378
4131.44132.6h1–3847
4231.34231.7h1–3916
4329.44330.9h1–4067
4417.34419.0m1–429
4514.54515.1h1–42100
 4074.44075.5h1–3710
JRF/AβN/25, GuHCl extract4132.14132.6h1–3840
4231.24231.7h1–3912
4330.34330.9h1–4059
4514.44515.1h1–42100
 3317.83318.9pyroGlu h11–428
 3336.93336.9h11–4225
JRF/cAβ42, GuHCl extract3356.03356.0m11–4212
4052.24052.6h5–426
4419.34419.0m1–428
4514.44515.1h1–42100
 3151.03152.7h11–408
JRF/cAβ40, GuHCl extract3170.23171.7m11–405
4328.24330.9h1–40100
4509.54515.1h1–429
JRF/rAβ/2, GuHCl extract4236.14234.8m1–4031
4423.44419.0m1–42100
4516.44515.1h1–4210

The MS spectrum of insoluble brain peptides immunoprecipitated with 4G8 was similar to that observed for soluble Aβ. However, we consistently observed two important differences, i.e. the signal for Aβ1–42 was more intense than for Aβ1–40, and there was a clear set of peaks for Aβ11–42 (Fig. 2d and Table 2). The increased peak size for Aβ1–42 was confirmed by IP/MS with JRF/AβN/25 (Fig. 2e and Table 2), and was in accordance with the ELISA data that demonstrated two- to threefold higher levels of Aβ1–42 compared to Aβ1–40 (Table 1). Detailed analysis of the Aβ11–42 peaks revealed that the middle peak corresponded to human Aβ11–42, the outer peak at slightly higher mass was identified as mouse Aβ11–42, and the smallest peak most likely represented the pyroglutamate (pyroGlu) derivative of human Aβ11–42 (Fig. 2d and Table 2). The same set of peaks for Aβ11–42 was present after IP with JRF/cAβ42/26, but not with any of the other antibodies used, confirming the suggested identity of the peptides (Table 2). Using either 4G8 or JRF/cAβ42/26, the peak intensity of mouse and human Aβ11–42 ranged in the order of 10% and 20%, respectively, of that of human Aβ1–42. Both antibodies also immunoprecipitated mouse Aβ1–42, the signal of which was approximately 10-fold lower than its human counterpart (Table 2), in accordance with our ELISA data. One additional peptide that was reproducibly picked up after IP/MS with JRF/cAβ42/26 had a mass corresponding to human Aβ5–42 (Table 2). IP/MS with JRF/cAβ40/10 yielded human Aβ1–40 as well as human and mouse Aβ11–40 (Table 2), but clearly the relative signal intensity of Aβ11–40 was much weaker than in the case of Aβ11–42.

IP/MS on GuHCl extracts of APPld mice with our rodent-specific antibody, JRF/rAβ/2, yielded a much more prominent signal for mouse Aβ1–42 than for Aβ1–40 (Fig. 2f and Table 2). This is again in good agreement with our ELISA data and suggests that mouse Aβ1–42 is preferentially deposited in plaques. Because the mouse 11–40/42 peptides lack the epitope to be immunoprecipitated with our antibody JRF/rAβ/2 (see Fig. 2a), the ratio of mouse Aβ1–42 versus Aβ11–42 could not be assessed from the MS spectrum. We reproducibly detected a weak signal for human Aβ1–42, but not human Aβ1–40 or Aβ1–38, in our IP/MS spectra, suggesting that the human Aβ1–42 peptide was coprecipitated with the mouse Aβ1–42 as dimer or oligomer.

Preferential deposition of human and mouse Aβ42 in APPld mice

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

Transgenic mice overexpressing human APP mutants have become an important tool in research on Alzheimer's disease. Here, we report on the characterization of APPld transgenic mice, leading to several important findings.

It was previously described that Aβ deposition in APPld mice starts at the age of 12–14 months, after which there is an exponential increase (Moechars et al. 1999). Unnoticed before, our data suggest that plaque formation is more extensive in females than in males. A similar observation was published for Tg2576 mice, which overexpress Swedish mutant APP (Callahan et al. 2001). Both transgenic mouse strains showed a plaque load in females that exceeded that in age-matched males by two- to three-fold. It was recently reported that this may be due to higher levels of synaptic zinc in female mice (Lee et al. 2002).

Both in males and females, the levels of human Aβ in brain deposits were two- to threefold higher for peptide 1–42 than for 1–40. This was demonstrated in this report by ELISA and IP/MS, and is in accordance with IHC studies that revealed preferential staining of the plaques in APPld mice with an antibody specific for Aβ42 (Moechars et al. 1999). The London and the Indiana mutations are known to increase the Aβ1–42 production relative to Aβ1–40, and consequently it is not surprising that we observed similar 42/40 ratios as reported for TgCRND8 (APP-KM670/NL671 + V717F, ratio 3.5) and PDAPP mice (APP-V717F mutation, ratio 7.8; Chishti et al. 2001). Interestingly, this feature clearly discriminates these transgenic mouse models from those that overexpress human APP only mutated at the Swedish site (Moechars et al. 1999; Kawarabayashi et al. 2001; Kuo et al. 2001), which deposit more Aβ40 than 42. In human AD patients, initial plaque formation is dominated by aggregation of Aβ42 (Iwatsubo et al. 1994; Wang et al. 1999), whereas at later stages Aβ40 also contributes significantly to the pool of ‘insoluble’ brain Aβ. Nonetheless, the ratio of Aβ42/Aβ40 is usually larger than 1 (Wang et al. 1999).

One of our most intriguing findings is that mouse Aβ is also deposited in the brains of 2-year-old APPld transgenic mice. This was shown by ELISA, IHC and IP/MS. Whereas ELISA and IHC can be subject to discussion on antibody specificity, our IP/MS data unambiguously demonstrate that mouse Aβ is indeed present in brain deposits. This indicates that human and mouse Aβ co-aggregate in plaques. We were surprised to find both by ELISA and IP/MS that the level of insoluble mouse Aβ1–42 was up to 10 times higher than that of the Aβ1–40 variant, and to our knowledge we are the first to report this. Analysis of soluble Aβ from brains of different transgenic or control mice (S. Pype, unpublished observations), or from culture medium of murine primary neurones (Cai et al. 2001), shows that there is more production of Aβ1–40 than Aβ1–42. Combined with our data this suggests that there is a strong preference for the longer peptide to aggregate in plaques.

Formation and deposition of Aβ variants in APPld mice

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

Our IP/MS analysis demonstrated that in addition to Aβ1–40 and 1–42, human Aβ1–38 was prominently present in the pool of soluble and insoluble Aβ peptides. This peptide was previously documented in samples of human AD brain (Kalback et al. 2002) and cerebrospinal fluid (Wiltfang et al. 2002) as well as extracts from APPsw transgenic mice (Kuo et al. 2001; Kalback et al. 2002) and PS1-APP double transgenic animals (McGowan et al. 1999). Although it cannot be ruled out that this Aβ variant is formed by proteolytic breakdown of Aβ1–40 or 1–42, recent findings suggested that cleavage at position 38 is due to a shift in γ-secretase activity (Weggen et al. 2001). We were surprised that the ratio of Aβ1–38 to Aβ1–40 is similar in extracts of soluble and insoluble peptides, indicating that the less hydrophobic Aβ1–38 (lacking two valines compared to Aβ1–40) is incorporated into plaques equally efficient. We also used IP/MS to look for variants longer than Aβ1–42, but were unable to detect any specific peak that could suggest the presence of Aβ peptides larger than 5 kDa.

Comparison of IP/MS spectra from soluble and insoluble peptides also revealed that short N-terminal Aβ variants (e.g. Aβ1–17, 1–19 and 1–20) were only detected as soluble peptides. These peptides may in part be generated by proteolysis of Aβ, but our own results from tissue culture experiments show that some of these Aβ variants are also formed when γ-secretase is inhibited (Vandermeeren et al. 2001), indicating that they are likely to be the products of alternative processing of APP-carboxyl-terminal fragment β1 (CTFβ1) by proteases like α-secretase or BACE2. Our data revealed that these peptides are not incorporated into plaques, which is not surprising in light of their hydrophilic nature.

Contrary to these findings, we observed selective deposition of Aβ11–42, and to some extent also Aβ1–42. The latter peptide was obviously present in the pool of soluble peptides, even at relatively high levels, but compared to Aβ1–40, the 42-residue variant was clearly enriched in plaques. Aβ11–42 is known to contribute significantly to total Aβ levels in elderly human brains (Naslund et al. 1994; Saido et al. 1996). We demonstrated that this truncated Aβ variant is also formed in transgenic animals. The fact that Aβ11–42 was not detected in soluble extracts, suggests that this hydrophobic peptide is strongly enriched in plaques. Cleavage of APP-CTFβ1 at position 11 is known to occur in vivo by a second β-secretase cut that precedes γ-secretase activity. Previous studies suggested that β-secretase processing at position 11 was species-specific (Cai et al. 2001), but our data show that human APP can be cleaved at Glu11 by the mouse β-secretase. A weak signal for Aβ cleaved at position 11 was observed in extracts from APP23 transgenic mice but not in other transgenic strains in which the Aβ composition was characterized by MS (McGowan et al. 1999; Kuo et al. 2001; Kalback et al. 2002). It is unclear whether this represents an actual phenotypic difference or whether it merely due to different experimental conditions.

Aβ deposits in the APPld brains also contained mouse Aβ11–42. Comparing peak intensities of mouse Aβ1–42 and Aβ11–42 on our 4G8-IP/MS spectra, it is possible that there is a similar amount of both Aβ variants incorporated into plaques. This would not be surprising because evidence is accumulating that rodent APP cleaved at position 11 is a major product of β-secretase activity (Buxbaum et al. 1998; Cai et al. 2001).

Human plaques contain substantial levels of pyroGlu derivatives of Aβ peptides starting at position Glu3 or Glu11 (Saido et al. 1996). Although we have no indication that the APPld mice produce Aβ starting at residue Glu3, the mice do have the pyroGlu Aβ11–42 variant present in the deposits. In PSAPP mice, Glu3 variants of Aβ were detected in amyloid deposits by IHC, but not by IP/MS with 4G8 (McGowan et al. 1999). Therefore it remains to be determined whether APPld mice contain this isoform at levels not detectable by IP/MS. In addition, the GuHCl extraction procedure that we used in this study may not solubilize all Aβ peptides. Although studies on Tg2576 and APP23 mice documented that amyloid deposits in transgenic animals are more soluble that human plaques (Kalback et al. 2002), we cannot exclude that further extraction with, e.g. formic acid would change the ratio's of Aβ variants seen in IP/MS or yield additional Aβ peptides.

The importance of Aβ variants in therapeutic use of AD models

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References

The characterization of transgenic AD models by IP/MS, as described here for APPld mice, is also of interest from a medical point of view. As passive immunization becomes a possible route for inhibition of Aβ deposition, the Aβ composition of the plaques could be of critical importance for the choice of antibodies, because clearance of all Aβ forms may be necessary for an optimal ‘therapeutic response’ in the mice. However, whether these observations can then be extrapolated from transgenic animals to AD patients remains to be shown. Our data corroborate the current insight that Aβ deposits in man and mouse differ in content, solubility and distribution. In addition, neurofibrillary degeneration, Lewy body pathology and cerebrovascular accidents are also very variable in their contributions to the dementia syndrome and it has to be seen to what extent a medical intervention that only affects amyloid deposition will affect these pathological lesions. These and other factors may have contributed to the recent failure of human vaccination trials and will require attention in the design of future clinical trials.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Preparation of mouse brain extracts
  5. ELISA quantification of soluble and deposited Aβ
  6. IP/MS analysis
  7. Immunohistochemical analysis of APPld brain slices
  8. Results
  9. Deposition of human and mouse Aβ in APPld mice
  10. Brains of APPld mice contain multiple Aβ variants
  11. Discussion
  12. Preferential deposition of human and mouse Aβ42 in APPld mice
  13. Formation and deposition of Aβ variants in APPld mice
  14. The importance of Aβ variants in therapeutic use of AD models
  15. Acknowledgements
  16. References
  • Buxbaum J. D., Thinakaran G., Koliatsos V., Ocallahan J., Slunt H. H., Price D. L. and Sisodia S. S. (1998) Alzheimer amyloid protein precursor in the rat hippocampus: transport and processing through the perforant path. J. Neurosci. 18, 96299637.
  • Cai H. B., Wang Y. S., McCarthy D., Wen H. J., Borchelt D. R., Price D. L. and Wong P. C. (2001) BACE1 is the major β-secretase for generation of Aβ peptides by neurons. Nat. Neurosci. 4, 233234.
  • Callahan M. J., Lipinski W. J., Bian F., Durham R. A., Pack A. and Walker L. C. (2001) Augmented senile plaque load in aged female β-amyloid precursor protein-transgenic mice. Am. J. Pathol. 158, 11731177.
  • Chishti M. A., Yang D. S., Janus C., Phinney A. L., Horne P., Pearson J., Strome R., Zuker N., Loukides J., French J., Turner S., Lozza G., Grilli M., Kunicki S., Morissette C., Paquette J., Gervais F., Bergeron C., Fraser P. E., Carlson G. A., George-Hyslop P. and Westaway D. (2001) Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein 695. J. Biol. Chem. 276, 2156221570.
  • Iwatsubo T., Odaka A., Suzuki N., Mizusawa H., Nukina N. and Ihara Y. (1994) Visualization of Aβ42(43) and Aβ40 in senile plaques with end- specific Aβ monoclonals: evidence that an initially deposited species is Aβ42(43). Neuron 13, 4553.
  • Janus C., Chishti M. A. and Westaway D. (2000) Transgenic mouse models of Alzheimer's disease. Biochim. Biophys. Acta 1502, 6375.
  • Kalback W., Watson M. D., Kokjohn T. A., Kuo Y. M., Weiss N., Luehrs D. C., Lopez J., Brune D., Sisodia S. S., Staufenbiel M., Emmerling M. and Roher A. E. (2002) APP transgenic mice Tg2576 accumulate Aβ peptides that are distinct from the chemically modified and insoluble peptides deposited in Alzheimer's disease senile plaques. Biochemistry 41, 922928.
  • Kawarabayashi T., Younkin L. H., Saido T. C., Shoji M., Ashe K. H. and Younkin S. G. (2001) Age-dependent changes in brain, CSF, and plasma amyloid β protein in the Tg2576 transgenic mouse model of Alzheimer's disease. J. Neurosci. 21, 372381.
  • Kuo Y. M., Kokjohn T. A., Beach T. G., Sue L. I., Brune D., Lopez J. C., Kalback W. M., Abramowski D., Sturchler-Pierrat C., Staufenbiel M. and Roher A. E. (2001) Comparative analysis of amyloid-β chemical structure and amyloid plaque morphology of transgenic mouse and Alzheimer's disease brains. J. Biol. Chem. 276, 1299112998.
  • Lee J. Y., Cole T. B., Palmiter R. D., Suh S. W. and Koh J. Y. (2002) Contribution by synaptic zinc to the gender-disparate plaque formation in human Swedish mutant APP transgenic mice. Proc. Natl Acad. Sci. USA 99, 77057710.
  • McGowan E., Sanders S., Iwatsubo T., Takeuchi A., Saido T., Zehr C., Yu X., Uljon S., Wang R., Mann D., Dickson D. and Duff K. (1999) Amyloid phenotype characterization of transgenic mice overexpressing both mutant amyloid precursor protein and mutant presenilin 1 transgenes. Neurobiol. Dis. 6, 231244.
  • Mercken M., Brepoels E., De Jong M., Laenen W., Raeymaekers P. and Geerts H. (2000) Specific ELISA systems for the detection of endogenous human and rodent Aβ40 and Aβ42. Neurobiol. Aging 21, S41.
  • Moechars D., Dewachter I., Lorent K., Reverse D., Baekelandt V., Naidu A., Tesseur I., Spittaels K., Van Den H. C., Checler F., Godaux E., Cordell B. and Van Leuven F. (1999) Early phenotypic changes in transgenic mice that overexpress different mutants of amyloid precursor protein in brain. J. Biol. Chem. 274, 64836492.
  • Naslund J., Schierhorn A., Hellman U., Lannfelt L., Roses A. D., Tjernberg L. O., Silberring J., Gandy S. E., Winblad B., Gard P. G., Nordstedt C. and Terenius L. (1994) Relative abundance of Alzheimer Aβ amyloid peptide variants in Alzheimer's disease and normal aging. Proc. Natl Acad. Sci. USA 91, 83788382.
  • Saido T. C., Yamao-Harigaya W., Iwatsubo T. and Kawashima S. (1996) Amino- and carboxyl-terminal heterogeneity of [β]-amyloid peptides deposited in human brain. Neurosi. Lett. 215, 173176.
  • Selkoe D. J. (1999) Translating cell biology into therapeutic advances in Alzheimer's disease. Nature 399, A23A31.
  • Van Dorpe J., Smeijers L., Dewachter I., Nuyens D., Spittaels K., Van Den H. C., Mercken M., Moechars D., Laenen I., Kuiperi C., Bruynseels K., Tesseur I., Loos R., Vanderstichele H., Checler F., Sciot R. and Van Leuven F. (2000) Prominent cerebral amyloid angiopathy in transgenic mice overexpressing the London mutant of human APP in neurons. Am. J. Pathol. 157, 12831298.
  • Vandermeeren M., Geraerts M., Pype S., Dillen L., Van Hove C. and Mercken M. (2001) The functional γ-secretase inhibitor prevents production of amyloid β1–34 in human and murine cell lines. Neurosci. Lett. 315, 145148.
  • Wang J., Dickson D. W., Trojanowski J. Q. and Lee V. M. Y. (1999) The levels of soluble versus insoluble brain Aβ distinguish Alzheimer's disease from normal and pathologic aging. Exp. Neurol. 158, 328337.
  • Weggen S., Eriksen J. L., Das P., Sagi S. A., Wang R., Pietrzik C. U., Findlay K. A., Smith T. W., Murphy M. P., Butler T., Kang D. E., Marquez-Sterling N., Golde T. E. and Koo E. H. (2001) A subset of NSAIDs lower amyloidogenic Aβ42 independently of cyclo-oxygenase activity. Nature 414, 212216.
  • Wiltfang J., Esselmann H., Bibl M., Smirnov A., Otto M., Paul S., Schmidt B., Klafki H. W., Maler M., Dyrks T., Bienert M., Beyermann M., Ruther E. and Kornhuber J. (2002) Highly conserved and disease-specific patterns of carboxyterminally truncated Aβ peptides 1–37/38/39 in addition to 1–40/42 in Alzheimer's disease and in patients with chronic neuroinflammation. J. Neurochem. 81, 481496.