| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,
,,,||,1
* Michael Smith Laboratories, University of British Columbia, Vancouver, BC, Canada;
Biomedical Research Centre, University of British Columbia, Vancouver, Canada;
Department of Medical Genetics, University of British Columbia, Vancouver, Canada;
Department of Microbiology & Immunology, University of British Columbia, Vancouver, Canada; and
|| Department of Zoology, University of British Columbia, Vancouver, Canada
1Correspondence: The Michael Smith Laboratories, 2185 East Mall, Vancouver, BC, Canada, V6T 1Z4. E-mail: wilf{at}brc.ubc.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: brain • microcirculation • vaccination • plaques
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6
7
8)
. It remains unclear how peripheral immunization can affect plaque burden in so-called "immunoprivileged" tissues, like the brain, that possess vasculature that limit immune surveillance. The blood-brain barrier (BBB) is formed by a continuous layer of capillary endothelium joined by tight junctions that are generally impermeable, except by active transport, to most large molecules including antibodies and other proteins. However, there have been no studies to date that have examined the integrity of the BBB before and after immunization. We previously showed there is increased permeability in the BBB of Tg2576 AD mice compared with age-matched controls at 10 months of age, as the signs of the disease become manifest (9)
. Moreover, the increase in BBB permeability was evident as early as 4 months of age, prior to disease onset and plaque deposition (9)
. Therefore, we hypothesize that disruption of the BBB appears to be another pathological hallmark of AD that may explain the mechanism by which immunization therapies have proved successful. The present study addresses this hypothesis by examining the effect of Aß immunization on the integrity of the BBB and thereby establishes a new paradigm for understanding the mechanism of action of Aß immunization in AD. |
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 11)
, under control of the hamster prion protein promoter (Charles River, Wilmington, MA, USA). Mice were maintained by mating Tg2576 males to C57B6/SJL F1 females and wild-type littermates were used as controls. Mice were fed lab chow adlibitum and kept under a 12 h light/dark cycle. All animal procedures were conducted with approval by the University of British Columbia Animal ethics committee.
Horseradish peroxidase injection
The Tg2576 and nontransgenic mice received injections of 15 mg of horseradish peroxidase (HRP; Sigma) dissolved in 0.25 mL of PBS into the tail vein. After 30 min, the mice were anesthetized with ketamine (25 mg/kg i.p.) and xylazine (5 mg/kg i.p.) and perfused with PBS, followed by 4% paraformaldehyde through the left ventricle of the heart. The brains were dissected, 100 µm sections were cut using a Vibratome, and extravascular HRP was visualized with 3,3'-diaminobenzidine (DAB; Vector Laboratories Inc.).
Aß vaccination protocol
The vaccination protocol was similar to that of Schenk et al. (1)
. Briefly, prior to immunization each mouse was bled and serum collected. Two groups of mice were vaccinated beginning at 6 wk and 11 months of age and sacrificed at 12 months and 15 months, respectively. Aß peptide was freshly prepared from lyophilized powder for each set of injections. For immunizations, 2 mg of Aß1-40 (human Aß1-40; Bachem) was added to 0.9 mL of deionized water and mixed until a uniform suspension was obtained. Then 100 µL of 10x PBS was added, and the solution was vortexed and placed at 37°C overnight until use the next day. Aß1-40 (100 µg antigen per injection) or PBS (control) was mixed 1:1 (v/v) with complete Freund's adjuvant for the first immunization. This was followed by a boost with Aß1-40 (100 µg) or PBS mixed 1:1 (v/v) with incomplete Freund's adjuvant at 2 wk and monthly thereafter. The 6-wk-old mice were vaccinated for a total of 11 months, and the 11-month-old mice were vaccinated for a total of 4 months. Aß or PBS alone was injected from the fifth immunization onward. Antibody titers for anti-Aß antibodies was assayed 2 wk after the initial injection by serial dilutions of sera against aggregated Aß, which had been coated in microtiter wells. Detection was done by using goat anti-mouse immunoglobulin conjugated to HRP and 2'2-AZINO-bis (3-ethylbenzthiazoline-6-sulfonic acid (ABTS; Sigma) as substrate. An absorbance plate reader (Spectra Max 190; Molecular Devices) then measured absorbance at 405 nm.
Evans blue assay
Quantitative Evans blue analysis was performed as described by Ujiie et al. (9)
. In brief, Tg2576 mice vaccinated with either Aß or PBS alone and their age-matched controls were weighed and injected i.p. with 50 µg/g Evans blue dye in PBS. Three hours after injection, the mice were anesthetized and perfused with PBS for 5 min. After perfusion, the brains were dissected and olfactory and cerebellum were removed, weighed, and Dounce homogenized in 0.5 mL of 50% trichloroacetic acid. The sample was then centrifuged at 13,000 rpm for 10 min. The supernatant was collected and diluted 1:4 in 100% ethanol and absorbance (620 nm) was measured using an ELISA plate reader. Values are graphed as OD620/brain weight, and the data were statistically analyzed with the 2-tailed Student’s t test, with P >0.05 as a significant cutoff value. As a control for Evans blue distribution throughout the mouse, OD620 values for the liver, a tissue known to be highly permeable, were also evaluated. All mice used in this study exhibited high levels of Evans blue dye in the liver.
Immunohistochemistry
Cerebral cortices of immunized and control mice were immunostained for the detection of amyloid deposits and activated microglia. Brains were fixed in 4% parafomaldehyde, paraffin embedded, and serial sections were cut at 8 µm. Sections were then deparaffinized, hydrated, and antigens were unmasked using DAKO Target Retrieval Solution (DakoCytomatiom) and 3% H2O2. The slides were rinsed and blocked, followed by an overnight incubation in either a mouse anti-human Aß monoclonal antibody, clone 4G8 (1:500, Signet Labs Inc.), or with a rat anti-mouse monoclonal antibody against microglia, F4/80 (1:10, Serotec). The slides were then washed and incubated with a secondary biotinylated anti-mouse antibody for 4G8 or a biotinylated anti-rat antibody for F4/80 (DakoCytomation) for 25 min, developed using DAB (Vector Laboratories Inc.), counterstained with Meyer’s hematoxylin, dehydrated, and mounted. Slides were examined under a Zeiss microscope and images were captured using OpenLab software. The number of plaques per neocortex (1)
brain section per mouse were counted and analyzed. Data were collected from four equally spaced sections. The values for all sections from one mouse were averaged to obtain a single sample for statistical analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. To assess the immune response of the mice to immunization with Aß, antibody titers for antibodies against Aß were measured. Serological analysis of serum samples collected from transgenic; nontransgenic mice were vaccinated with Aß and PBS was analyzed for the titers of anti-Aß antibodies by ELISA using synthetic Aß1-40 peptide. Serum samples from mice in all groups were collected after the second injection and results indicated that transgenic and nontransgenic mice vaccinated with Aß produced a high IgG response to Aß1-40. No detectable antibody titers were observes in transgenic and nontransgenic mice vaccinated with PBS (Fig. 1
a, b). Mice vaccinated with Aß, which did not exhibit high anti-Aß antibody titers, were not used in the study.
|
Decrease in amyloid plaques and activated microglia after Aß immunization
Next, amyloid plaque burden was assessed in both Aß and PBS immunized transgenic and nontransgenic animals. As noted in other immunization studies (1
, 2
, 4
, 8
, 12
13
14)
, there was a significant reduction in the plaque burden in the cortex and hippocampus of transgenic mice immunized with Aß compared with those immunized with PBS. In 15-month-old mice immunized with Aß after disease onset, there was a decrease in plaque burden with a reduction in both the size and number of plaques, whereas brain sections from PBS-treated transgenic mice contained numerous amyloid deposits (Fig. 2
a, b; Fig. 3
a). These data agree with previous studies where a dramatic decrease but not total elimination or prevention of plaques was observed in these animals. In mice immunized with Aß prior to disease onset (12 month), there was an almost complete prevention of Aß deposition. In Aß immunized transgenic mice, four of six had no detectable amyloid deposits. Two mice from this treatment group had a single isolated plaque in the four brain sections examined. As with the 15-month group of mice, 12-month-old, PBS-treated transgenic mice exhibited numerous amyloid deposits in their cortical and hippocampal regions (Fig. 2c, d
; Fig. 3b
). There were no detectable plaques in nontransgenic controls vaccinated with either Aß or PBS. This confirms that immunization with Aß may prevent plaque formation and thereby protect against disease.
|
|
The presence of activated microglia in immunized and nonimmunized animals was analyzed. Activated microglia are usually found associated with the amyloid plaques and can be distinguished from resting microglia by the expression of specific proteins, such as IL-1ß, CD11b, and major histocompatibility class II, and by distinct morphology (15)
. Activated microglia exhibit an altered morphology from resting microglia, first by the presence of thickened ramified processes and larger cell bodies, which progresses to a final amoeboid/phagocytic state (16
, 17)
. Sections of mouse brains were reacted with an antibody specific for the F4/80 antigen. The F4/80 antigen is expressed by a majority of mature macrophages, including microglia (18)
. In agreement with previous studies (1)
, there appeared to be a reduction in the presence of activated microglia in the brains of Aß immunized mice compared with PBS controls, suggesting a dampening of the inflammatory response. In 15 month transgenic mice immunized with PBS, there are more plaque infiltrating microglia present than in those immunized with Aß (Fig. 4
e, g). There were more densely stained F4/80 positive microglia with swelled cell bodies and thickened processes in PBS-treated transgenic mice compared with mice treated with Aß. Similarly, in 12-month-old transgenic mice immunized with PBS there were more amoeboid shaped and thick ramified microglia, whereas the microglia in the Aß immunized mice have smaller cell bodies and more extensively ramified processes (Fig. 4f, h
). The microglia present in nontransgenic mice displayed largely nonactivated morphology with small cell bodies and thin, highly branched processes (Fig. 4a-d
). Therefore, postimmunization with Aß reduced the number of activated microglia.
|
The BBB is compromised in Tg2576 AD model mice
We addressed the integrity of the BBB in unimmunized Tg2576 AD mice by perfusing mice using a tracer, HRP, or fluoresceinated BSA commonly used techniques to visualize vessel leakage. As depicted in Fig. 5
a, b, HRP can be detected in the brain parenchyma of transgenic mice whereas no HRP is visualized in nontransgenic controls, indicating that the BBB in permeable in the transgenic mice. Similar results we demonstrated for BSA (supplemental data).
|
BBB permeability decreases after Aß immunization
To test the effect of Aß vaccination on the BBB, permeability of the barrier was assessed in the 15 and 12 month Tg2576 mice immunized with Aß or PBS. BBB permeability was assessed by the standard quantitative Evans blue assay (19
, 20)
. Organs with a selective barrier, such as the brain, only take up a small amount of the dye. If there is a breach in the BBB, then Evans blue, like HRP or BSA, will enter the parenchyma, whereas it is excluded from the brain with an intact BBB.
In both groups of mice (15 and 12 month), the integrity of the BBB of transgenic mice was compromised compared with that of nontransgenic controls (Fig. 6
; **P<0.05). These data concur with the study by Ujiie et al., where BBB integrity was compromised in 4- and 10-month-old Tg2576 mice (9)
and with the data presented here showing the presence of HRP in the brains of transgenic mice and no BSA or HRP in the brains of transgenic mice. Upon immunization with Aß, the AD transgenic mice displayed a significant decrease in BBB permeability, indicating a dramatic restoration of the integrity of the BBB. Transgenic mice injected with Aß had a significantly lower amount of Evans blue dye in their brain parenchyma compared with AD transgenic mice injected with PBS alone (Fig. 6
; *P<0.05). This was true for both 15-month-old mice and 12-month-old mice immunized with Aß. In the 12-month-old mice, it is possible that vaccination with Aß maintained an intact BBB, with little to no compromise in BBB function. There was no change in BBB permeability in nontransgenic mice injected with Aß or PBS. This latter observation is interesting as Aß has been shown to be cytotoxic to endothelial cells and neuron in vitro by eliciting apoptosis (21
22
23)
and angiogenesis (24)
, yet the low amounts used in this study do not appear to affect BBB integrity in normal mice.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, including AD. BBB disturbances have been associated with stroke (25)
, cerebrovascular ischemia (25)
, hypertension (26)
as well as with mutations in the apolipoprotein gene (20)
, all of which are risk factors for AD. Consequences of a faulty BBB can lead to the leakage of various neurotoxic substances into the brain, resulting in neuronal damage and plaque formation (27)
. The present study demonstrates that the BBB is compromised in AD and that immunization with Aß repairs the damage found in BBB in Tg2576 AD mice. The observation that the BBB of Tg2576 AD mice is compromised compared with control littermates is illustrated by the presence of HRP and BSA in the brain parenchyma (Fig. 5
and supplemental data). The extent to which the BBB is compromised appears to be minimal, since acute pathologies such as microinfarcts were not readily visible, but this damage remains of great consequence to the brain and appears to be cumulative over time. However, we have demonstrated that immunization with Aß peptides reverses BBB permeability. The significant decrease in Evans blue uptake observed in Aß immunized transgenic mice compared with the PBS mice suggest that Aß immunization results in the complete restoration of the BBB. More intriguing and exciting is the observation that in mice inoculated prior to disease onset, Aß immunization appears to be able to actually prevent further disease progression. In these mice, Evans blue uptake is comparable to nontransgenic mice, where there is no measurable decreased integrity of the BBB (Fig. 6)
. One possible mechanistic explanation of the restoration of the BBB in older mice is that the vaccination leads to the decrease in the amount of circulating Aß that could directly or indirectly affect the function of endothelium in the BBB (28)
. It has been shown that the effect of widespread Aß deposition is degeneration and death of endothelial cells and the obliteration of the capillary lumen (29)
. Moreover, ultrastructural studies indicate that 
32% of fibrillar amyloid plaques are in contact with one or more cerebral capillaries (30)
and that 77% of plaques in the Tg2576 mice and 91% of human plaques are in direct contact with capillaries (30)
.
As demonstrated in many studies (1
2
3
, 8
, 31)
, there is a significant decrease in Aß deposits and microgliosis after immunization (Figs. 2
, 4)
. With the removal of Aß from the brain and the subsequent deactivation of microglia, there will be a decrease in the brain in the amount of reactive oxygen and nitrogen compounds and inflammatory cytokines produced by these cells, which have been shown to activate vascular endothelial cells (32
33
34)
.
The relationship between amyloid and endothelial cells has recently been under extensive investigation. Aß has been demonstrated to elicit many proapoptotic and proangiogenic responses in the endothelial cells that make up the BBB (35
, 36)
. It has been exhibited in vitro that exposure of endothelial cells to µM concentrations (5–25 µM) elicits proapoptotic signals (21
, 24
, 37)
whereas treatment with nM concentrations (50–250 nM) of Aß elicits proinflammatory signals and increased monocyte migration with minimal disruptions to the endothelial monolayer (38
, 39)
. Treatment of primary cerebral mouse endothelial cells with Aß25-35 resulted in the activation of AP-1 and the subsequent expression of Bim, a member of the BH3 only family of proapoptotic proteins (37)
. Moreover, Aß treatment also resulted in the translocation of second-mitochondria derived activator of caspase (Smac), a regulator of apoptosis, from the mitochondria to the cytosol, where it can bind to the X chromosome-linked inhibitor of apoptosis protein (XIAP), resulting in cell death (37)
. Cytochrome C release from the mitochondria and the subsequent activation of several caspases, such as caspase 8 and caspase 3, all important events on the apoptotic cascade, have been reported (40)
. Moreover, angiogenesis is also observed in endothelial cells in AD. For example, inflammatory mediators such as TNF-
, IL-1ß, and IL-6 that stimulate angiogenesis (41)
, COX-2, and amyloid have been found to cause an increase in the expression of many angiogenic factors including VEGF, TGF-ß, and TNF- (42
, 43)
. Furthermore, in AD patients there is an increase in VEGF expression in the brain as well as an increase in the serum and CSF (42
, 44)
. Therefore, it is proposed that in response to proapoptotic or proangiogenic signals, tight junctions are altered; cell rounds up creating a permeable barrier. We hypothesize that with the removal of signal provided by Aß, the endothelial cells quiesce and reform tight junctions, thereby recreating a tight barrier. More studies are needed to study the exact nature of the tight junctions as the disease progresses in this mouse model, as well as what happens as a consequence of immunization.
Few mechanisms have been proposed to explain the success of Aß vaccination. In one scenario, antibodies to Aß enter the brain, bind to Aß in the plaques, and dissociate the plaque. These Aß-antibody complexes are then degraded by microglia via FcR-mediated or FcR-independent phagocytosis, or the entire complex is sequestered in the brain and not able to form new plaques. Alternatively, antibodies to Aß bind to peripheral Aß causing disequilibrium between plasma and brain amyloid. Therefore, an efflux of Aß from the brain via plaque dissociation can occur in order to reestablish equilibrium. Our data may help link both of the proposed models. Studies have shown that immunoglobulins reactive with Aß are present in the serum, CSF, and amyloid plaques of AD patients as well as with amyloid-laden vessels (45
46
47)
. With a permeable BBB, antibodies are likely able to gain access into the brain, bind to Aß, and are either degraded by microglia via FcR-mediated or independent phagocytosis or the Aß -antibody complex can leave the brain. Paradoxically, recent studies suggest that FcR-mediated phagocytosis appears not to be required for antibody-mediated clearance of plaques (48)
. These data suggest that another method of antibody-mediated clearance must be involved. By corollary, with a leaky BBB, Aß may be able to move from the brain and CSF to the plasma more freely, allowing catabolism in peripheral tissues. Finally, human clinical trials of Aß immunization in patients with AD were halted because adverse effects were noted in some patients (29
, 49
, 50)
. Aseptic meningoencepalitis was detected in a small portion of patients, and autopsy of one patient revealed inflammatory infiltration, multiple cortical hemorrhages, and severe small cerebral blood vessel disease (29)
. Though this was not noted in the present study with the Tg2576 AD mice, a report by Racke et al. (28)
demonstrates that upon immunization of the PDAPP transgenic mice with Aß, cerebral amyloid angiopathy-associated microhemorrhage results. Thus, although it is clear from our study that the integrity of the BBB is restored, it remains to be demonstrated whether all endothelium are devoid of pathology. In summary, these observations provide a fundamentally new intellectual framework for understanding the efficiency of vaccination to modify disease formation. More provocatively, our data suggest that reestablishing the integrity of the BBB provides a new intervention point for modifying disease outcome in amyliodopathies such as AD.
Received for publication May 2, 2005. Accepted for publication October 5, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Schroeter, K. Khan, R. Barbour, M. Doan, M. Chen, T. Guido, D. Gill, G. Basi, D. Schenk, P. Seubert, et al. Immunotherapy Reduces Vascular Amyloid-{beta} in PDAPP Mice J. Neurosci., July 2, 2008; 28(27): 6787 - 6793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Paul, S. Strickland, and J. P. Melchor Fibrin deposition accelerates neurovascular damage and neuroinflammation in mouse models of Alzheimer's disease J. Exp. Med., August 6, 2007; 204(8): 1999 - 2008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. V. Nikolic, Y. Bai, D. Obregon, H. Hou, T. Mori, J. Zeng, J. Ehrhart, R. D. Shytle, B. Giunta, D. Morgan, et al. Transcutaneous beta-amyloid immunization reduces cerebral beta-amyloid deposits without T cell infiltration and microhemorrhage PNAS, February 13, 2007; 104(7): 2507 - 2512. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
