Complement Regulation-Piet Eikelenboom
Complement C1 and C3 have been implicated in maintaining the solubility of Aβ. As far back as early 1982, Eikelenboom's group showed that the main source of complement was neurons and not glia. It is the microglia that generate PI cytokines such as IL-1β which induce neurons to make βAPP and complement; the neuron is just not an innocent bystander in PI pathways in AD brain. Therapeutic intervention to augment complement activation may be best aimed at serine proteases C1r/s (C1 subcomponents) which would block C1 activation before C3, the major effector protein of the complement cascade, is activated.
The Role of Complement Anaphylatoxin C5a in Neurodegeneration-Implications for AD-Giulio Pasinetti.
Mice deficient in C5 complement are more susceptible to hippocampal excitotoxic lesions. C5-derived anaphylatoxin C5a may protect against excitotoxicity both in vitro and in vivo. Mechanisms of C5a-mediated neuroprotection include MAP kinase activation. In transgenic C5a receptor knockout mouse (C5aR-KO), alteration of C5aR alters calcium/calmodulin kinase signal transduction detrimental to normal brain cell function. In AD, an increase in cyclooxygenase-2 (COX-2) expression has been reported at CDR 0.5-1.0, right around the time of the first signs of Aβ plaque density increases. COX-2 presence and elevations are a key marker for AD onset.
α1-ACT, α2-MAC, Serum Amyloid P-Ikuo Tooyama.
Acute phase reactants are a heterogeneous collection of proteins upregulated in blood plasma when PI responses occur. These are chiefly α1- antichymotrypsin (ACT), α2 macroglobulin (α2 MAC) and serum amyloid P (SAP). α1-ACT is a serine protease inhibitor involved in PI, strongly binds to Aβ and stimulates its polymerization (bad stuff!), and also contributes to the resistance to proteolytic degradation. α2 MAC is also a potent proteinase inhibitor-also binds to Aβ and prevents its degradation (more bad stuff!!). In AD brain, α2-MAC may also act as a chaperone for Aβ and prevent Aβ clearance. SAP, a serum plasma protein that is present in all SP deposits, binds complement proteins C4 and C1. The general upregulation of such acute phase proteins cumulatively is bad news for AD brain.
Fibrinolysis/coagulation systems-Haruhiko Akiyama.
AD lesions inducing PI cascades resemble chronic inflammation in the periphery. Just as it functions in the periphery, thrombin in the brain causes rapid retraction of neuronal and astrocyte processes. Fibrinolysis is involved in neurite outgrowth, neuronal migration, neuronal plasticity and excitotoxicity. Secreted APP (sAPP) is identical to protease nexin-2 (down-regulated in AD). Nexin-2 inhibits coagulation, indicating a physiological role of the intrinsic coagulation pathway in AD brain. Coagulative process increased in AD brain-it can be found in all mature SP.
Cyclooxygenase-1, Cyclooxygenase-2-Kerry O'Banion
McGeer et al. (1990) were some of the first to point out that AD prevalence in patients with rheumatoid arthritis (taking NSAIDs ) was 6-12 times less frequent then expected. Rogers et al. (1993) first showed AD patients treated with indomethacin showed reduced progression of AD. Since COX's were one of the known targets for NSAID action, these data both suggested the probable involvment of inducible cyclooxygenases (COX-2) enzymes in AD brain. A key marker for AD brain is increased IL-1β, verified by many groups. Upregulation of IL-1β is a super potent inducer of COX-2 gene expression. Aβ doesn't increase COX-2 by itself but IL-1β and Aβ show synergistic induction of COX-2 in AD brain. COX-2, as the key enzyme in PI mediator activation is elevated in AD brain and is the target of several anti-PI pharamacological studies. These include the specific COX-2 inhibitors NS-398 (Taisho), Celcoxib (Searle) and Rofecoxib (Merck). Large trials for AD are currently under way and preliminary results are promising.
IL-1/S100β-W.S.T. Griffin, U. Arkansas, USA.
Griffin and coworkers have shown that IL-1β: (1) is overexpressed by microglia in AD brain; (2) is overexpressed in areas with the most Aβ deposition; (3) induces excessive synthesis and processing of βAPP in neurons; (4) overexpression correlates with the transition from fibrous Aβ to condensed amyloid senile plaques (SP) and (5) strongly activates astrocytes. IL-1β converting enzyme (ICE), also known as caspase 1, the enzyme that converts pro-IL-1β to mature functional IL-1β, is also up regulated in AD. Overexpression of ICE contributes to DNA damage observed in neurons adjacent to SP. IL-1 may have a more direct function in cholinergic dysfunction. Overexpression of IL-1 induces expression of acetylcholinesterase both in vitro and in vivo. Finally, specific polymorphisms in several IL-1 genes increase risk for AD.
IL-6 is a pleotrophic cytokine that mediates immune responses and PI reactions affecting cell growth and differentiation. Primary IL-6 synthesis is in human brain microglia and astrocytes. It is a major pyrogen and increases vascular permeability and lymphocyte activation-all of which have been shown to be elevated in AD brain. IL-6 RNA message and protein increases rapidly as little as 30 minutes after prostaglandin E2 (PGE2) treatment of astrocytes. Prostainoids (COX-2 end products) such as PGE2 induce astrocytic production of IL-6. NSAIDs inhibit COX-2 but do not overtly inhibit IL-6 synthesis in astrocyte cultures. More specific inhibitors of IL-6 are one important and attractive targets for AD therapy in the future.
Transforming growth factor β's (TGFβ's) are multifunctional polypeptide growth factors which play prominent roles in tissue development homeostasis and repair and are expressed in all brain cells-neurons, astrocytes, and microglia. Just like for other cytokines, for TGFβ, too much of a good thing is a bad thing. TGFβ is also widely found in SP cores, in CSF and serum of AD cases. TGF can be a strong promoter of PI mechanisms. Cross-breeding of TGF transgenic mice with hAPP mice promote deposition of Aβ in cerebral blood vessels rather than in plaques. Some subforms of TGFβ (β1 and β2) are shown to protect against neuronal injury by inhibition of apoptosis through induction of Bcl-2 and stabilization of intracellular calcium.
Summary-Nancy Ruddle, Yale University.
1. Most if not all PI mediators are increased in AD brain and associated with classical neuropathological structures such as the SP.
2. Cyclic pattern of the PI response in AD brain is positively reinforcing, so stopping anywhere around the cycle might be expected to stop the feed-forward mechanism (ideally).
3. Don't blame everything on the glia/astroctyes. Neurons can do a lot of damage to themselves and to glia.
8 July 2000. Microglia, or "The Natural and Un-Natural History of Brain Microglia."-Mark Emmerling
Microglia are the resident macrophages of the CNS, make up ~10% of total CNS cells and perform many functions of peripheral macrophages including migration, phagocytosis, and production of immune molecules like cytokines and chemokines. Microglia may be a good pathological marker for AD due to the following reasons: (1) Natural aging shows an elevated microglial activation response which in AD is exacerbated (see Streit, Progress in Neurobiol 1999 57:563-581). The difference between "resting" microglia and "activated" microglia is phenomenal. This can raise microglial IL-1β RNA message production by a factor of 1,000 to 10,000 (!); (2) Aβ, and deposition into SP are the key pathological markers for microglial activation. Aβ induces membrane perturbation through membrane-binding sites for Aβ (the serpin enzyme complex receptor), C3b, C5 activation products, heparin sulfate proteoglycans, LDL receptor related proteins, RAGE, Aβ binding alcohol dehydrogenase (ABAD)-all strongly PI. Interaction of Aβ with microglia ends up generating more IL-1α, IL-1β, etc. Microglial targeted treatments should focus on NF-kB antagonists and PPAR c agonists (e.g., propentofyline, a potent anti-PI compound).
Activated astrocyte products are invariable components of SP in AD and suggest an instigating or proliferative role. Astrocyte products in AD include trophic molecules, adhesion molecules, cytokine S100β-all PI associated proteins. S100β normally promotes neurite outgrowth, maintains free calcium levels and when dysregulated induces neuronal βAPP gene expression, IL-6 and NOS activities. A strong correlation exists between S100β-expressing astrocyte density and SP number. IL-1β, shown by many reports to be elevated in AD induces S100β, αACT, ApoE and C3 as part of an agressive "cytokine cycle" in AD brain.
βAPP and Aβ are the key PI markers for AD. Their lab sees an induction of the inducible prostanoid generating COX-2 RNA and protein (isozyme) in global ischemia models. They also observe a dramatic induction of the proinflammatory transcription factors (TF) AP1 and NF-kB, but not of TF Oct1 or SP1 in iscehmia. Glutamate increases NF-kB-DNA binding only in mixed neuron-glia cultures suggesting neuron-glia crosstalk in initiating PI pathways. A neuron specific NF-kB in rat cultures they called NK-BF for neuronal K B factor. NK-BF may be rat-specific and show only rat-specific responses in PI signaling. sAPP, the "good" amyloid derived from APP is generally thought of as neuroprotective, but can induce IL-1 after further "abnormal" sAPP processing in PI brain cells. The sAPP N-terminus appears to be PI, while the sAPP COOH terminus appears to be neurotrophic (impications for α-, β-, or c- and other secretase activities in contributing to PI in AD brain).
Aβ Binding Proteins-Scott Webster.
Aβ binding proteins (1) modulate formation of Aβ and SP; (2) affect induction of Aβ mediated PI pathways and neurotoxicity; (3) influence Aβ clearance mechanisms such as the susceptibility of Aβ to proteolysis; (4) influence Aβ as a target for phagocytosis; (5) modify Aβ-mediated stimulation of PI-mediator production; and (6) alters βAPP gene expression pathways. C1q, SAP and ACT are amyloidogenic. ApoE and α2 MAC are anti-amyloidogenic.
Oxidative Stress-Douglas Walker.
Free radicals are constantly produced by cells as the result of oxidative metabolism. The brain has a high metabolic rate so there are lots of reactive oxygen species (ROS) to deal with. When clearance of ROS is compromised, oxidative stress is increased as appears to be the case in AD. Aβ peptides can prime ROS production in rat microglia (Klegeris et al., 1997). ROS can also induce phagocytosis of glia and IL-1β release. Aβ at 10 uM induces RAGE (receptor for advanced glycation end-products). Markers of oxidative stress in AD brains include elevated advanced glycation endroducts, malondialdehydes, elevated 4-OH nonenol (indicated hyperoxidized lipids), elevated carbonyl (oxidized proteins), nitrotyrosine modified proteins, etc. These changes are also seen in transgenic mice that are βAPP overexpressors. Antioxidant therapy may be beneficial in dealing with ROS in AD brain.
RAGE and the Macrophage Scavenger Receptor-David Stern.
RAGE is a multiligand member of the IG superfamily of cell surface molecules and contributes to cell oxidative stress accompanying amyloidosis. Amyloid interaction with cell surface RAGE promotes accumulation of Aβ fibrillation. RAGE is down-regulated during human brain development and is up-regulated in pathological states just like basal levels of COX-2. The RAGE promoter contains multiple NF-kB-DNA binding sites and the Tfs cINF and NF-IL6 are important regulators of RAGE transcription. RAGE activation increases production of macrophage colony stimulating factor (MCSF) and IL-6-anti-RAGE antibodies block this induction. When neurons bind Aβ they also produce MCSF. MCSF activates microglia to produce more IL-1, and the PI cycle increases by a feed forward mechanism. Inhibiting key components of the RAGE pathways may be useful therapeutically.
"The lesions of AD are characterized by the presence of a host of PI molecules." Chronic inflammation is involved in AD, heart attack (MI), and stroke, which take three of four people prematurely. NSAIDs reduce the risk of MI 25%-85%, AD 50%-75%, stroke, colon and breast cancer, and cataract 25%-50%. McGeer referred to the flak he and Joe Rogers took 15 years ago when they first used the term "neuroinflammation"-then considered an oxymoron-which is now of mainstream AD interest. The concepts of PI processes in human brain were reviewed historically, from Metchnikoff (1843-1916) who first described phagocytes, a local PI process and Hortega, who first suggested that brain microglia gathered as phagocytes during brain injury, to Ralf von Furth who described microglial activation in the brain (and could never get funded) to Jules Bordet (1870-1961) who described brain complement, rubor (cytokine induction) and also said inflammation in local. In summary: AD is a mixture of autoimmune and autotoxic processes. Inflammatory outcomes are (1) healing, (2) death, or (3) chronic inflammation, which is the stalemate situation observed in AD. Note that antiinflammatory compounds (aspirin) are the most prevalent class of drugs ever developed by man. NFT involvement in the induction of PI cascades was briefly disussed. Both senile plaques and NFT are PI inducers. To date, NSAIDs are the preferred alleviator molecules for quenching PI pathways.
Choosing the right drug and the right time for AD neuroinflammation therapy-John Breitner.
The AD anti-inflammatory prevention trial (ADAPT) is now funded and will assess efficacy of the conventional COX1/COX2 inhibitor naproxen and celecoxib (selective COX2 inhibitor) for primary prevention of age-related and AD cognitive decline. Trials of hormone replacement therapy and histamine H2 blockers are attractive (no major side effects). The aim is to prevent onset of AD by retarding the progression of the disease in its latent stages. The compounds lose effectiveness in later stages. Short trials to date have been to date discouraging.
Amyloid Vaccination and Its Neuroimmunologic Implications-Dale Shenk.
Aβ immunization of PDAPP mice works remarkably well in preventing Aβ deposits from forming and in reducing progression of plaque deposits and neuropathology in older animals. Therefore, a directed immune response against Aβ can be beneficial in reducing AD-like pathology.
Neuroimmunology of Aging-Caleb Finch.
Normal aging involves subsets of the same PI phenomenon as in AD, and AD is an "accumulated aging" response. In normal aging, there are increased deposits of various tissue amyloids, but they differ from Aβ in the brain. AD is part of a broad "gero-inflammatory" response fueled by the oxidative processes associated with aging.