Posted 5 December 2005

E-mail discussion Printable version

Return to Discussion Text

Ben Barres
Hi, June!

June Kinoshita
Hi, everyone!

Wenan Qiang
I'm Wenan Qiang; I work at Linda Van Eldik's lab.

Ben Barres
Hi, Wenan; what do you work on?

Wenan Qiang
I am working on glia inflammation.

Ben Barres
That's great!

Linda Van Eldik
Hi, everyone.

Ben Barres
Linda, do you work on AD? Wenan, do you mean microglia or reactive astrocytes?

Gabrielle Strobel
Could everyone identify themselves briefly as we begin, so we all know who's here. I am Gabrielle Strobel, managing editor of the Alzforum.

June Kinoshita
June Kinoshita, executive editor of Alzforum.

Ben Barres
Hi, I’m Ben Barres; I am a neurobiologist at Stanford interested in neuron-glial interactions.

Linda Van Eldik
Linda Van Eldik, Northwestern University, co-director of the Center for Drug Discovery and Chemical Biology, associate director of the Northwestern Alzheimer's Disease Center, studying mechanisms of glial activation and the potential of targeting chronic glial activation for AD therapeutics.

Chiara Cupidi
I'm Chiara Cupidi; I work at University of Verona, Italy.

Gabrielle Strobel
Ben, Linda does beautiful work on glial inflammation and the role of ApoE, among other players, in that. Ben, do you think aging is the strongest risk factor for late-onset AD because of the astrogliosis observed in the aging brain?

Ben Barres
Great question! I have read that with aging some neurons are slowly dying, just from old age, and anytime neurons die, reactive gliosis seems to occur. So there definitely is probably more reactive gliosis with aging.

Gabrielle Strobel
Tuck Finch has data on that showing correlates with some performance drops in aging, but for AD I wonder if astrogliosis could make the brain more vulnerable to other insults?

Ben Barres
Gabrielle, yes, it's another great question. What is the point of astrocytes being reactive—is it good or bad? Perhaps it could be reversed, but maybe that would have adverse consequences. For instance, Mike Sofroniew showed that reactive astrocytes are important in sealing up the blood-brain barrier (BBB) after brain injury (see Bush et al., 1999).

Gabrielle Strobel
And I believe Tony Wyss-Coray has data on active astrocytes taking up amyloid (see ARF related news story).

Ben Barres
Do you mean that Tony has found that reactive astrocytes are better at taking up amyloid than are regular astrocytes?

Gabrielle Strobel
Ben, do you think reactive gliosis could slow the normal stream of lactate from glia to neurons and thus starve them? I believe it's known that degenerating neurons run low on adenosine-5'-triphosphate (ATP) before they die, and there are many studies on a "hypometabolic" state of degenerating neurons.

Wenan Qiang
Glia activation is a cellular response to brain injury. In chronic glial activation, astrocytes will lose their neuronal support function. This is, therefore, detrimental to neurons.

Ben Barres
Wenan, Mike Sofroniew’s paper shows that reactive gliosis is important in keeping neurons alive after injury. If he blocks gliosis, the neurons die at a much higher rate. But it is not clear if that is related to their BBB function or to neurotrophic support of neurons, or other mechanisms.... Gabrielle, there has been a great deal of work on division of metabolic labor between astrocytes and neurons. All that work has been done with normal astrocytes. I am not aware of studies where people have addressed whether reactive astrocytes share metabolic pathways with normal astrocytes or not. It seems like that would be an important avenue of investigation, given the points you raise.

Linda Van Eldik
I think the key is to think of it as a delicate balance. Glia activation is probably generally beneficial, such as in acute brain injuries. But when it is chronic, as in AD, the detrimental responses may outweigh the beneficial responses, and then it has deleterious consequences to the brain.

Ben Barres
What happens to ApoE expression/secretion in reactive gliosis, Linda?

Linda Van Eldik
Ben, ApoE expression is upregulated by Aβ in activated astrocytes, at least in culture.

Gabrielle Strobel
Linda, is that "because" the astrocytes crank up their ability to clear Aβ via ApoE?

Linda Van Eldik
Gabrielle, I'm not sure of the mechanism, but we published a paper with Mary Jo LaDu showing that the Aβ stimulation of ApoE production requires the LDL receptor (LaDu et al., 2001).

Ben Barres
Linda, thanks for the information. I didn't know! Does ApoE bind to Aβ? I am just wondering if Aβ is a normal signaling molecule (at low levels) and whether ApoE helps deliver that signal to other cells?

Gabrielle Strobel
Ben, I believe it does. Dave Holtzman's hypothesis is that ApoE helps clear Aβ, and when it is not properly lipidated, or is in the ApoE4 isoform, it does a poor job of that and instead promotes deposition. Correct me if that's wrong, Linda.

Linda Van Eldik
ApoE definitely binds Aβ, with E3 binding better than E4.

Wenan Qiang
Does that mean Aβ metabolism requires ApoE?

Gabrielle Strobel
Wenan, I believe that is the implication. Last week, three papers in the Journal of Biological Chemistry reported that loss of ATP-binding cassette transporter, subfamily A (ABCA), which shuttles out cholesterol, reduces ApoE greatly but does not lower amyloid deposition in AD mouse models (see ARF related news story).

Linda Van Eldik
Aβ is a signaling molecule. It can bind to various cellular receptors and stimulate signal transduction pathways in the glia (at least in astrocytes and microglia) and in the neurons. It stimulates glia activation responses and it stimulates neuronal dysfunction/death.

Ben Barres
Linda, can you clarify? Are there known receptors for Aβ? (Sorry if that's ancient history.) What about the role of clusterin/ApoJ in all this? Does ApoJ also bind to Aβ? I ask because I am really impressed by how high the levels of ApoJ are in astrocytes in vivo (based on our unpublished gene chip data). Come to think of it, I do recall an old paper of Jerry Silver’s where he shows profound effects of Aβ on the phenotype of astrocytes (see, for example, Canning et al., 1993). I wonder what receptors mediate that.

Linda Van Eldik
ApoJ also binds Aβ. Tuck Finch and Grant Krafft/Bill Klein published papers showing that ApoJ can affect Aβ's ability to polymerize. Clusterin (ApoJ) alters the aggregation of amyloid-β peptide (Aβ1-42) and forms slowly sedimenting Aβ complexes that cause oxidative stress (see Oda et al., 1995). It promotes oligomeric forms of Aβ, and thereby enhances toxicity. Aβ can bind to a number of different receptors, including receptor for advanced glycation end products (RAGE), scavenger receptors, a receptor complex described by Gary Landreth (see Bamberger et al., 2003)....

Ben Barres
Has anyone looked at whether Aβ binding to ApoJ or ApoE enhances or hinders receptor interactions with lipoprotein receptor-related protein (LRP), etc.?

Gabrielle Strobel
Ben, Dave Holtzman also had a paper, I think in PNAS, about knocking down ApoJ in AβPP transgenics and looking at effect on amyloid. I don’t recall the phenotype, but can send you the citation (see DeMattos et al., 2002). Linda probably knows better than I.

Linda Van Eldik
Mary Jo LaDu also has some data relevant to that issue (see, for example, LaDu et al., 2000 reporting that ApoE mediates Aβ effects on astrocytes; Manelli et al., 2004 on effects of ApoE isoform on Aβ interactions; Tokuda et al., 2000 on the effect of lipidation on ApoE/Aβ interactions.

Ben Barres
Linda, so I guess the question is whether activation of any of those receptors would be sufficient to mediate known Aβ signaling effects like the induction of reactive gliosis; it seems unlikely to me. Has anyone taken tagged Aβ (perhaps in oligomeric form) and screened a cDNA library yet for receptors?

Gabrielle Strobel
I don't know.

Ben Barres
I saw a paper from Klein showing that Aβ oligomers bound to synapses, but I don't think he looked at whether there was binding to glia, as well (see Lacor et al., 2004).

Gabrielle Strobel
Yes, I seem to recall he saw punctate binding on cultured neurons and then tried to isolate what he thinks is a receptor complex.

Ben Barres
Gabrielle, that sounds pretty interesting! I don't see how the field can move forward in understanding Aβ actions without elucidating the receptors that mediate its effects. I am particularly interested in whether Aβ serves a normal signaling role, for instance, in the normal developing or normal adult brain. But I don't know if there is any work yet that pertains to this.

Linda Van Eldik
Take a look at our paper in 2000 (see LaDu et al., 2000). We found if we blocked ApoE receptors, then the ability of Aβ to stimulate IL-1b and ApoE production in glia were blocked. The mechanism is so far unknown.

Ben Barres
Linda, that's very interesting—I didn't know! Which ApoE receptors are expressed by astrocytes? Is it known?

Linda Van Eldik
We haven't looked for glia receptors, but oligomeric Aβ definitely can stimulate glia activation. We did a study with Mary Jo LaDu showing differential effects of oligomeric versus fibrillar Aβ on glia-induced inflammatory responses. Oligomeric Aβ is an excellent activator of glia (see White et al., 2005).

Ben Barres
We have a good rat brain cDNA library, Linda, if you would like to screen it for glial receptors!

Linda Van Eldik
Astrocytes express at least low-density lipoprotein receptor (LDLR) and LRP, but I'm not sure about any others.

Ben Barres
Linda, I will check our data and let you know.

June Kinoshita
I'm curious about aging-related changes in glia themselves. How does aging affect the normal function of glia, in particular, their role in maintaining healthy neuronal functions, and also, how does aging alter the reactive response of glia?

Ben Barres
June, I know that reactive gliosis is not a robust property of neonatal or early postnatal brain. I believe that developing astrocytes only gain the ability to become reactive when they finish dividing and maturing about P14 (in rodents, I mean). Quite possibly, this is because immature astrocytes are very similar (or the same) as reactive astrocytes, but that is a total guess.

Wenan Qiang
June, cellular redox balance setup is changing along with aging.

Ben Barres
June, but I think those are great questions. Never thought about them before and am not aware of relevant work. I also think that it would be very interesting to know the extent to which aging affects the number of normal glia. No one’s looked at that, to my knowledge.

Gabrielle Strobel
Are there studies comparing the transcriptional profile of normal versus activated astrocytes, or young versus old? Given that astrocytes influence neurons in many different ways—metabolic, neurotrophic signals, cholesterol/ApoE, etc. —it might be interesting to know what changes as they become increasingly activated, or simply as they age.

Ben Barres
We have just been doing some gene chip work, Gabrielle, transcriptional profiling of astrocytes at various stages of development. We now have the ability to compare reactive astrocytes with the normal astrocytes, and those experiments are on the drawing board. There is a postdoc joining the lab soon who hopes to work on this.

Gabrielle Strobel
Ben, to follow up on an earlier point: The only two physiological roles for Aβ (not AβPP) that I am aware of are modulating synaptic activity and regulating the levels of cholesterol and sphingomyelin. I am not aware of a physiological signal transduction role that has good data.

Ben Barres
However, I am sure that in terms of aging it will be extremely difficult to isolate astrocytes from aging brain. But I've been seeing some very nice work with laser capture from specific cell types in human brain, so it might be possible to do this for aging astrocytes in sections of human brain. For instance, see the beautiful work of Arnon Rosenthal looking at gene expression by human dopaminergic neurons (see, for example, Grimm et al., 2004).

Gabrielle Strobel
Ben, that sounds like a very interesting project. Is it possible to include some astrocytes from disease states?

Ben Barres
Gabrielle, my understanding of the work by Malinow is that he did not address whether Aβ has a normal physiological role in a normal brain, but only in tissue in which it had been overexpressed. It would be fantastic if he would look at whether it has a similar role in normal brain (as you say).

Gabrielle Strobel
Ben, I agree. The work on Aβ regulating cholesterol sphingomyelin synthesis, however, is done with physiological levels of Aβ. It's by Tobias Hartmann and just came out in Nature Cell biology last week (see ARF related news story).

Linda Van Eldik
Joe Rogers’ lab has developed a nice method for culturing glia from AD brain samples with short postmortem intervals.

Ben Barres
Linda, I am very skeptical about studies of astrocytes cultured from brain these days because we are finding that most astrocytes instantly die in culture and that the few that survive are either glial progenitors (which persist in adult brain) or newly generated immature glia from glial progenitors. To be sure, there are many similarities among all the astrocyte stages, but this is a caveat.

Linda Van Eldik
Ben, all cultured cells have some caveats associated with them. They are good for doing certain studies, like defining mechanisms, signaling, etc. However, we also have data from the mouse where infusion of oligomeric Aβ induces a robust glia activation and induction of proinflammatory cytokines like IL-1b, TNFα, and S100B (see Kim et al., 2004).

Ben Barres
Linda, that's very interesting. I had somehow missed the paper. In our studies of normal astrocytes we do not see any expression of TNFα or most of the usual cytokines, even in development, which may suggest that reactive astrocytes are quite different from immature astrocytes.

Linda Van Eldik
Ben, actually, I think the IL-1 and TNF are being produced by the microglia, and the S100B by the astrocytes. We haven't done colocalization studies in the mouse brain tissue, but did in glial cultures.

Ben Barres
Linda, that makes sense! Thanks for the reference! Though I am concerned that the cell types that survive from aging brain may actually be a different cell type—antigenic identification is important. I am always puzzled about the relationship between microglial activation and reactive astrocytosis—which comes first?

Linda Van Eldik
Ben, it depends on which endpoints you are measuring. The TNFα is one of the earliest cytokines to go up, IL-1 next, then iNOS and S100B in astrocytes. But there is a spectrum of responses occurring at different times after the stimulus. So it's not easy to say one comes first.

June Kinoshita
Ben, you point out in the background text that astrocytes and oligodendrocytes express more AβPP than do neurons. Has anyone looked at the effects of familial AD (FAD) mutations or of AβPP overexpression in astrocytes and oligodendrocytes?

Ben Barres
June, a very interesting question. I think so far the focus has been on the neurons, although I know there has been previous work showing that astrocytes secrete enormous amounts of certain AβPP splice forms like sAβPP. What these splice forms might do seems understudied, as well. I think I've seen some stuff about possible trophic effects. One problem is that some of these proteins are heavily glycosylated, which may be important to their function, but in most studies recombinant forms that are not properly modified are studied.

Gabrielle Strobel
Ben, yes, sAβPPα is usually regarded as a neurotrophic form, the "good" product of α-secretase cleavage, and many studies in the field are now focusing on ways to shift AβPP cleavage away from a BACE-γ pathway and toward this α pathway.

Chiara Cupidi
What about the role of glia in white matter? In corticobasal degeneration, in multiple system atrophy (MSA), and in AD, too, pathological proteins are often deposited there. Can these proteins create axonal dysfunction or lead to neuronal damage by retrograde or anterograde transport?

Gabrielle Strobel
Chiara, a great question. John Trojanowski and Virginia Lee just had a J. Neurosci paper (see Higuchi et al., 2005) about this, saying that tau filaments form in oligodendrocytes in animal models of cerebellar and brainstem atrophy (CBA) and frontotemporal dementia with parkinsonism (FTDP). And they also proposed that the toxic effect on neurodegeneration goes via axonal transport. We are also planning a live discussion with Phil Landfield, which will focus on oligodendroglia in neurodegeneration.

June Kinoshita
Ben, going back to your comment about isolating astrocytes from aging brain, can you comment about what the difficulties are? I've also often wondered whether anyone has attempted to count astrocytes in postmortem brain, comparing normal and AD tissues. Is this technically really formidable?

Ben Barres
June, counting astrocytes in brain, say, in AD and in normal aging, would require much better astrocyte markers than exist today. We have just identified, through our gene chip studies, some better markers and are working with a company to make antibodies available quickly to everybody.

June Kinoshita
That sounds great! Let us know when you've got them, and we'll add them to our antibody database.

Linda Van Eldik
Astrocytes can be isolated from aging brain, but they don't survive as well in culture and grow very slowly. Tuck Finch, Joe Rogers, Sue Griffin, and others have cultured astrocytes from aged rodents and from AD brain.

Gabrielle Strobel
Why are there so many leads but few solid data on glia that are then widely accepted? Are the big hurdles more due to lack of interest or technical challenges? Lack of reliable and specific glial markers appears to be one.

Ben Barres
Gabrielle, I guess there are so many hard technical challenges to studying glia, starting with the lack of good markers and the lack of good culture models for mature astrocytes (that will change soon, as we've made some interesting progress on that).

Gabrielle Strobel
What do you think is the function of the glial AβPP you see expressed so abundantly? Hui Zheng at Baylor has data on the role of AβPP in synapse formation (synapses are defective without it), but I am not aware if she distinguishes between neuronal or glial AβPP (see Yang et al., 2005).

Ben Barres
Gabrielle, yes, I saw the very interesting work of Zheng at Baylor. And Vivian Budnik just had a paper on the role of AβPP in synaptogenesis in flies (see Ashley et al., 2005 and ARF related news story). I hadn't thought of the point you raise about how they can distinguish between glia and neuronal forms, which certainly is a good question, given the recent work on the role of Schwann cells in synaptogenesis at the neuromuscular junction.

Gabrielle Strobel
Yes, we've reached the end of the hour. I certainly have a lot of other questions, but we'll follow up on the topic in future chats. Apologies for the software hiccups, and thank you all for coming.

Ben Barres
Before everyone goes, I just want to say this has been a great chat for me—I've learned a lot! Thank you, all!

Gabrielle Strobel
I have to go but want to stress that you can all stay and continue the conversation as long as you like. Good bye for now.

Linda Van Eldik
Well, I must go, too. Just want everyone to know that targeting neuroinflammation selectively has great potential for AD therapy in the future. Perhaps that will be the subject of a future chat. Bye!

Chiara Cupidi
Bye!