We assessed whether peripheral activation of microglia by a nasal proteosome-based adjuvant (Protollin) that has been given safely to humans can prevent amyloid deposition in young mice and affect amyloid deposition and memory function in old mice with a large amyloid load.
Amyloid precursor protein (APP) transgenic (Tg) J20 mice received nasal treatment with Protollin weekly for 8 months beginning at age 5 months. Twenty-four-month-old J20 mice were treated weekly for 6 weeks.
We found reduction in the level of fibrillar amyloid (93%), insoluble β-amyloid (Aβ; 68%), and soluble Aβ (45%) fragments in 14-month-old mice treated with Protollin beginning at age 5 months. Twenty-four-month-old mice treated with nasal Protollin for 6 weeks had decreased soluble and insoluble Aβ (1-40) and (1-42) and improved memory function. Activated microglia (CD11b+ cells) colocalized with Aβ fibrils in the 24-month-old animals, and microglial activation correlated with the decrease in Aβ. No microglial activation was observed in 14-month-old mice, suggesting that once Aβ is cleared, there is downregulation of microglial activation. Both groups had reduction in astrocytosis. Protollin was observed in the nasal cavity and cervical lymph node but not in the brain. Activated CD11b+SRA+ (scavenger receptor A) cells were found in blood and cervical lymph node and increased interleukin-10 in cervical lymph node. No toxicity was associated with treatment.
Our results demonstrate a novel antibody-independent immunotherapy for both prevention and treatment of Alzheimer's disease that is mediated by peripheral activation of microglia with no apparent toxicity. Ann Neurol 2008;63:591–601
There is increasing evidence that activation of microglial cells is associated with amyloid clearance in transgenic (Tg) mouse models.1–6 For example, multiphoton microscopy shows amyloid clearance by anti–β-amyloid (Aβ) antibody is associated with increased microglia activation,7 which may relate to FcR-mediated phagocytosis of Aβ immune complexes by microglia. Thus, cellular mechanisms that enhance microglia phagocytosis of Aβ could play an important role in the immunotherapy of Alzheimer's disease (AD).8
We previously found that nasal vaccination with a proteosome-based adjuvant (Protollin), comprising purified outer membrane proteins of Neisseria meningitides and lipopolysaccharide that is well tolerated in humans plus glatiramer acetate (GA), a US Food and Drug Administration–approved synthetic copolymer used to treat multiple sclerosis, decreases Aβ plaques in an AD mouse model.6 This effect did not require antibody, as it was observed in B-cell–deficient mice. We now report that Protollin alone prevents accumulation of Aβ given chronically at an early stage of amyloid deposition and also reduces amyloid when given to older animals.
Materials and Methods
(B6XDBA)F1 J20 APP Tg mice express a mutant form of the human amyloid protein precursor bearing both the Swedish (K670N/M671L) and the Indiana (V717F) mutations (APPSwInd).6 Mice were housed in a pathogen-free facility in accordance with Harvard guidelines and approved by the Harvard Medical School institutional animal care and use committee.
Protollin9 was obtained from Glaxo Smith Kline (Laval, Quebec, Canada). GA (Copaxone)10 was obtained from the Brigham and Women's Hospital pharmacy (Boston, MA).
Protollin (1μg/mouse) was given on days 1, 3, and 5 of the first week followed by a weekly boost. Glatiramer (25μg/mouse) acetate with or without Protollin was given on an identical schedule. Mice received a weekly boost beginning at age 5 months until age 14 months. When treatment was given with Prollolin in 24-month-old mice, the weekly boost was given for 6 weeks. Phosphate-buffered saline (PBS) or bovine serum albumin (BSA) was used as controls. In prior experiments, we found no effect of nasal PBS or BSA on amyloid load in APP Tg mice.6
Amyloid load was measured as total Aβ by using enzyme-linked immunosorbent assay6 and as amyloid fibril using thioflavin-S staining.6 The right hemisphere of each mouse in each treatment group was homogenized with PBS contain protease inhibitor and centrifugated at 75,000g for 30 minutes to quantify total Aβ. The supernatant-containing soluble Aβ was stored at −70°C. The pellet containing insoluble Aβ was extracted in 5.0M guanidinium-chloride (pH 8) for 3 hours at room temperature. Dilutions were used to measure levels of Aβ1-x by enzyme-linked immunosorbent assay.11
Lymphocyte Cell Culture
Proliferation and cytokine measurements were done as described previously.6
Microglia/macrophages (CD11b+, MCA74G) was supplied by Serotec, Bicester, United Kingdom and astrocyte (glial fibrillary acidic protein–positive) Sigma, St. Louis, MO and by Rabbit anti-amyloid antibodies (R1282) was a gift from Dennis Selkoe. Quantification was done as previously described.6 For pathological evaluation of liver, lung, kidney, and brain, hematoxylin and eosin staining (six animals per group) and pathological evaluation were done in a fashion blinded to the treatment.
Mice were anesthetized using isoflurane and immobilized in a stereotaxic apparatus. One injection of 1μg/2μl Protollin or PBS was delivered over a 2-minute period into each hippocampus as described previously.3 Stereotaxic coordinates from bregma were −1.8mm posterior, ±1.4mm lateral, and −2.0mm ventral. The experiment consisted of five animals. Each animal received Protollin injection to one hemisphere and PBS to the other hemisphere. The mice were killed at 2 days.
Tissue from the right hippocampus or cervical lymphoid tissue was analyzed for messenger RNA expression via reverse transcription followed by real-time polymerase chain reaction (PCR) using TaqMan. Reverse transcriptase PCR assays were designed by Applied Biosystems (Foster City, CA) as described previously.12
Data comparisons were performed using Student's t test when two groups were compared or one-way analysis of variance when three or more groups were analyzed.
Prevention of Amyloid Deposition in Amyloid Precursor Protein Transgenic Mice by Treatment with Nasal Protollin Beginning at 5 Months of Age
We reported that nasal Protollin and GA reduce amyloid burden when given for 6 weeks in 14-month-old APP Tg mice.6 To test this treatment as prevention, we treated age- and sex-matched littermates from APP J20 Tg weekly beginning at 5 months of age with GA, Protollin, GA+Protollin, or PBS and killed at age 14 months. We found a reduction of insoluble Aβ (68%; p < 0.002) and amyloid fibril (93%; p < 0.001) in nasal Protollin versus control animals that received nasal BSA (Fig 1). Adding GA to the Protollin did not improve the effect, and no effect of nasal GA alone was observed (Table 1). We also found a significant (45%; p < 0.05) reduction in soluble brain Aβ by enzyme-linked immunosorbent assay in Protollin-treated animals (see Fig 1). The reduction of soluble brain Aβ was associated with an increase in soluble serum Aβ (35%; p < 0.05). No effect of GA on soluble Aβ was observed (see Table 1).
Table 1. Effect of Nasal Protollin on Total and Fibrillar β-Amyloid Given for 8 Months in Amyloid Precursor Protein Transgenic Mice
Prevention of Amyloid Deposition by Chronic Nasal Protollin Is Associated with Decreased Astrocytosis and No Evidence of Toxicity
Astrocytosis occurs under many circumstances including in response to neuronal damage. We found a marked decrease in astrocytosis in the hippocampus of Protollin-treated animals (11% in control animals vs 2.1% in treated animals; p = 0.001) as measured by glial fibrillary acidic protein staining (see Fig 1). We observed a slight increase in microglial activation (from 79 to 90 cells/mm2) that was not statistically significant. Histological analysis of liver, lung, kidney, and brains of Protollin-treated animals after 8 months of treatment showed no toxicity. Protollin-treated animals exhibited no toxicity as measured by body weight, eating habits, tail tone, or mobility. We also assessed mobility at 8 months using open-field testing, an index of locomotor/exploratory activity and anxiety in rodents, and found no impairment in Protollin-treated animals versus non-Tg littermates or 5-month-old APP Tg mice.
Nasal Protollin Decreases Amyloid Burden in 24-Month-Old APP Transgenic Mice
To test nasal Protollin in old APP Tg animals with significant amyloid deposition, we administered nasal Protollin or BSA to 24-month-old mice weekly for 6 weeks and measured brain and hippocampus. Nasal Protollin decreased soluble Aβ (1-40) and (1-42) by 36 and 38% and insoluble Aβ (1-40) and (1-42) by 56 and 79%, respectively, in the brain (Fig 2 and Table 2). There was also reduced Aβ burden in the hippocampus of Protollin-treated mice (p < 0.05) (see Fig 2B). No effect was observed in the percentage of brain area containing thioflavin-S–positive amyloid fibrils in the hippocampus. We also found improved cognition as measured by fear conditioning (see Supplementary Fig. 1).
Table 2. Effect of Nasal Protollin on Total and Fibrillar Aβ in 24-Month-Old Amyloid Precursor Protein Transgenic Mice
Activated Microglia Colocalize with β-Amyloid Deposition and Correlate with β-Amyloid Clearance in 24-Month-Old Amyloid Precursor Protein Transgenic Mice Treated with Nasal Protollin
We performed immunohistochemical analyses to assess microglial activation and scavenger receptor expression in the brains of nasal Protollin-treated mice in relation to Aβ deposition in the hippocampus. Microglial cells and macrophages express scavenger receptor A (SRA), also called macrophage scavenger receptor 1 (MSR1), which can bind Aβ that accumulates in the brain and cerebral blood vessels in AD,13 and promote endocytosis of Aβ and adhesion to Aβ-coated surfaces.14 We found colocalization of SRA+ microglia with amyloid deposits in the hippocampus (Fig 3). We then quantified microglia activation by the number of SRA and CD11b+ cells per square millimeter in the hippocampal region, and found an increase in SRA+microglia and CD11b+ cells in Protollin-treated animals versus control (p < 0.001) (see Fig 3B). Because Protollin is given nasally and may stimulate CD11b+ cells in the periphery, we examined expression of CD14 (lipopolysaccharide receptor) and SRA (MSR1) on CD11b+ cells from cervical lymph node. We found an increased percentage of CD11b+SRA+ and CD11b+CD14+ cells in the cervical lymphoid tissue after 5 weeks of nasal Protollin (p < 0.002) (see Fig 3C). We then tested the blood of Protollin-treated animals and found an increase of CD11b+SRA+ cells (p < 0.05) (see Fig 3D), suggesting that SRA+ macrophages are activated in the periphery and migrate to the brain where they play a role in the clearance of Aβ. We did not find increased T-cell responses in the spleen to Aβ after nasal Protollin as measured by cytokines or proliferation (control = 3,005 ± 1,082cpm; Protollin = 2,850 ± 1,102cpm; background counts, 100–300cpm). However, after anti-CD3 stimulation, we found an increase of interleukin-10 (IL-10; 1,800 ± 120pg/ml) versus control (720 ± 63pg/ml) (p < 0.001), suggesting an antiinflammatory response after Protollin treatment. IL-10 level increase was also observed in the cervical lymph nodes (Fig 4) after nasal Protollin and when Protollin was directly injected into the brain (see Fig 6).
Nasal Protollin Activates CD11b+ Cells in Cervical Lymphoid Tissue
APP Tg mice were treated nasally with Protollin or PBS every other day for 5 days and killed 1 hour after the last treatment. By immunohistochemistry using a Protollin-specific antibody, we found Protollin in the nasal cavity and cervical lymph nodes of treated animals. In the cervical lymph node, Protollin colocalized with CD11b+ cells (see Fig 4A). Protollin could not be detected in the brain. Protollin could also not be detected in the olfactory bulb or brains of Balb/c mice after one or two nasal administrations of 1μg (not shown). Balb/c mice are the most sensitive for observing nasal retrograde transport into the olfactory bulb and brain. Real-time PCR of cervical lymph nodes demonstrated increases of toll-like receptor-4 (TLR-4) and C-C-chemokine receptor-2 (CCR2), a receptor for CC chemokine ligand 2 (CCL2) chemokines. CCR2 has been shown to play an important role in migration of cells to the central nervous system and in reduction of amyloid load.15 We also found increased expression of the antiinflammatory cytokine IL-10 and significant increase of insulin-degrading enzyme (IDE), which is involved in amyloid degradation.16
Reduction in Amyloid Levels and Increase in Interleukin-10 after Intrahippocampal Protollin in Amyloid Precursor Protein Transgenic Mice
Even though Protollin appears to work peripherally, to further characterize the biological properties of Protollin, we injected Protollin or PBS intrahippocampally into the left or right hemisphere, respectively of 16-month-old APP Tg mice to determine the type of inflammatory response and cytokine patterns after direct injection. After 2 days, there was reduced Aβ compared with the left hemisphere or PBS injection (p < 0.05), and this was associated with activation of CD11b+ cells (Fig 5). Protollin colocalized with CD11b+ cells at the site where the Protollin was injected (Fig 6A). Real-time PCR from the hippocampus demonstrated an increase in microglia markers (CD11b, SRA, TLR-4, TLR-2), chemokine CCL3 that plays an important role in migration of microglia,8 and IDE (see Fig 6). Investigation of cytokine expression showed an increase in IL-10 and a decrease in IL-12 and IL-23 levels, which is consistent with the increase in IL-10 we observed in the cervical lymph node after nasal Protollin administration. Thus, even when Protollin is injected directly into the brain, there is an increase in the antiinflammatory cytokine IL-10 and a decrease in the proinflammatory cytokines IL-12 and IL-23. No changes were observed in interferon-γ or transforming growth factor-β (not shown).
We investigated an antibody-independent approach to activate microglia and reduce Aβ in a mouse model of AD in a prevention paradigm in which treatment was begun at age 5 months, before the deposition of amyloid. We found that a proteosome-based adjuvant given nasally activated microglial-like cells expressing SRA and prevented accumulation of amyloid.
Previously, we reported that nasal Protollin plus GA given for 6 weeks decreased Aβ plaques in 14-month-old APP Tg mice, and the extent of microglial activation correlated with the decrease in Aβ.6 Reduction in Aβ load occurred with Protollin alone, though to a lesser degree. In this study, we have made three new observations. First, we observed no difference between nasal Protollin and nasal Protollin + GA in the prevention of Aβ accumulation when given weekly for 8 months to 5-month-old APP Tg mice. Second, we found an increase of CD11b+SRA+ cells in the blood of animals treated with Protollin, which provides a marker that can be used for human trails. Third, though we found a decrease in Aβ at 14 months after 8 weeks of Protollin treatment, there was no increase in activated microglia (CD11b+ cells) or T cells compared with control. We hypothesize these results relate to the fact that treatment was begun as early as 4 1/2 months when there is almost no sign of amyloid deposition or evidence of microglial activation. Thus, nasal Protollin clears Aβ just as it forms, and because microglial activation depends on Aβ deposition, once Aβ is cleared, there is reduction in microglial activation. Herber and colleagues17 reported that lipopolysaccharide injected into the brains of APP Tg mice transiently activated microglia and led to Aβ clearance after which the microglia reaction resolved. A similar effect was observed with direct injection of anti-Aβ antibody.5 This is in contrast with what we observed in 24-month-old animals treated for 6 weeks with nasal Protollin where we observed persistent microglial activation. Of note, Butovsky and colleagues18 demonstrated that subcutaneous GA without adjuvant can lead to migration of microglia-like cells from the periphery in old APP mice. In our work, nasal GA alone was not sufficient for amyloid clearance.
An important question is how our current results with Protollin alone given to young mice (4–5 months) chronically over 8 months relate to our previous studies in which Protollin plus GA decreased Aβ levels in old (14 months) APP Tg mice.6 In both studies, we observed an effect of Protollin alone, but in neither study did we observe an effect of GA alone. Thus, an adjuvant is required for nasal GA to have its effect. The addition of GA to nasal Protollin enhanced clearing Aβ in older mice (including fibrillar deposits), but Protollin alone was equally as effective when treatment began early when there was little or no amyloid deposition. Thus, it appears that GA plus Protollin is more effective at clearing Aβ when there is advanced disease and a large Aβ burden that includes fibrillar deposits. We believe the enhanced effectiveness of Protollin plus GA in older mice relates to activation of T cells by GA, which enhances microglia activation, and thus Aβ clearance. Both GA and Protollin have antiinflammatory properties that may serve to lessen side effects from the treatment. Specifically, GA induces Th2/Th3 responses, and Protollin increases IL-10 and decreases IL-12/IL-23 (see later discussion).
Because it is believed that a Th1-cell response against Aβ caused meningoencephalomyelitis in 6% of AD patients,19 shifting the balance from Th1-type responses may be important for preventing this potential complication. Furthermore, immunization with amyloid-β in adjuvant induced encephalitis in APP Tg mice in which a transgene encoding interferon-γ was expressed under the control of the astrocyte-specific myelin basic protein promoter.20 This suggests that the proinflammatory cytokine interferon-γ may be necessary to facilitate the induction of the Th1-cell responses to amyloid-β in the brain, and this, in part, may explain the occurrence of encephalitis in AD patients. Thus, an immune strategy that does not involve inducing reactivity to Aβ and that favors antiinflammatory (eg, IL-10, transforming growth factor-β) responses would be preferable.21 We found that nasal Protollin induced immune responses in the cervical lymph node characterized by increase in IL-10, and after intrahippocampal injection of Protollin we found an increase of IL-10 and decrease of the proinflammatory cytokines IL-12 and IL-23 in the brain. This suggests that nasal Protollin would be less likely to be associated with side effects in humans.
The increase of the major Aβ scavenger receptor on the surface of macrophage-like cells in the cervical lymphoid tissue and blood suggests the effect of Protollin related to the migration of these cells from the periphery to the brain. This is consistent with the fact that we did not find Protollin in the brain of nasal Protollin-treated animals. Furthermore, we did not detect damage to the blood–brain barrier as measured with fibrinogen (data not shown); this peripheral migration does not involve damage to vessel walls. Microglia may originate from the bone marrow and enter the circulation as monocytes. During the early embryonic stages, monocytes migrate through the blood–brain barrier into the brain parenchyma and differentiate gradually into microglia.22 Migration of blood-derived monocytes into the brains of Tg mouse models of AD exhibiting age-dependent Aβ deposition has been reported.23, 24 Those cells can then play an important role in reduction of amyloid load. In support of this, APP Tg mice deficient in CCR2 accumulate Aβ earlier and die prematurely, in a manner correlated with CCR2 gene dosage, indicating that absence of early microglial accumulation leads to decreased Aβ clearance.15 CCR2 is a chemokine receptor expressed on microglia, which mediates the accumulation of mononuclear phagocytes at sites of inflammation. CCR2 is also required for migration of CD11b+ cells into the brain, and we found increase of CCR2 in the cervical lymphoid tissue after nasal Protollin. In addition, we found upregulation of IDE, which has been associated with degradation of Aβ.16 The finding of activation of CD11b+SRA+ cells in the blood of animals treated with nasal Protollin provides a means to measure its immunological effects in humans.
We also tested nasal Protollin given weekly for 6 weeks in 24-month-old APP Tg mice to test its effect in animals with a large amyloid burden. We found a reduction in both soluble and insoluble forms of Aβ (both 1-40 and 1-42), though there was no significant reduction in fibrillar Aβ as measured by thioflavin-S. These results are consistent with our previous studies in which only GA plus Protollin was effective in clearing fibrillar Aβ in older animals, whereas Protollin alone reduced total Aβ. We also found improved cognition in 24-month-old animals treated with Protollin alone as measured by fear conditioning. Although our results do not suggest that Protollin acts directly in the brain, direct hippocampal injection demonstrated similar effects to what we observed in the cervical lymph node after nasal Protollin. Specifically, intrahippocampal injection was associated with the upregulation of the antiinflammatory cytokine IL-10, downregulation of the proinflammatory cytokines IL-12 and IL-23, upregulation of IDE, and upregulation of TLR-2 and TLR-4.
In summary, we demonstrate that nasal Protollin activates microglia/macrophage cells to clear amyloid without induction of an inflammatory signal that could lead to neurotoxicity, and when young animals are treated in a prevention paradigm, there is no residual microglial activation once the amyloid is cleared. Furthermore, nasal Protollin decreases astrocytosis in animals with a large amyloid burden. Despite the fact that we did not observe toxicity in animals given Protollin, we caution that long-term administration of Protollin in humans with AD might cause side effects that we did not observe in our animal studies. Nonetheless, because we found a reduction of Aβ without apparent toxicity when given weekly for 8 months and single doses of Protollin have been given to people without toxicity, we speculate that Protollin may be beneficial for both the prevention and treatment of AD.
This work was supported by the NIH (National Institute of Aging, H.L.W.), Alzheimer's Association (H.L.W., D.F.), the Gruss Lipper Foundation and HFSP (D.F.).
We thank Dr D. Selkoe for helpful discussions and critical review of the manuscript.